JOSÉ AUGUSTO SGARBI
ASPECTOS EPIDEMIOLÓGICOS DAS DISFUNÇÕES TIROIDIANAS
NA POPULAÇÃO NIPO-BRASILEIRA DE BAURU
Tese de Doutorado apresentada à Universidade Federal de
São Paulo – Escola Paulista de Medicina, para a obtenção do
título de Doutor em Ciências
São Paulo
2011
1
Sgarbi, José Augusto
ASPECTOS EPIDEMIOLÓGICOS DAS DISFUNÇÕES TIROIDIANAS
NA POPULAÇÃO NIPO-BRASILEIRA DE BAURU, São Paulo, 2011.
p.96.
Tese de Doutorado, Programa de Pós Graduação em Endocrinologia,
Disciplina de Endocrinologia, Departamento de Medicina, Escola Paulista
de Medicina, Universidade Federal de São Paulo
Orientador: Prof. Dr. Rui Monteiro de Barros Maciel
Co-orientadora: Profa. Dra. Luiza K. Matsumura
Descritores: thyroid diseases, thyroid diseases, epidemiology, mortality
2
JOSÉ AUGUSTO SGARBI
ASPECTOS EPIDEMIOLÓGICOS DAS DISFUNÇÕES TIROIDIANAS
NA POPULAÇÃO NIPO-BRASILEIRA DE BAURU
Tese de Doutorado apresentada à Universidade
Federal de São Paulo – Escola Paulista de Medicina,
para a obtenção do título de Doutor em Ciências.
Programa de Pós-Graduação em Endocrinologia
Clínica.
Orientador:
Prof. Dr. Rui Monteiro de Barros Maciel
Co-orientadora:
Profa. Dra. Luiza K. Matsumura
Coordenador do Programa de Pós-Graduação:
Prof. Dr. Sérgio A. Dib
Pró-Reitor de Pesquisa e Pós-Graduação:
Prof. Dr. Arnaldo Lopes Colombo
São Paulo, 2011
3
Banca Examinadora
1. Orientador: Rui Monteiro de Barros Maciel, Professor Titular, Disciplina de
Endocrinologia, Departamento de Medicina, Escola Paulista de Medicina, Universidade
Federal de São Paulo (UNIFESP).
Membros:
2. Antonio
Roberto
Chacra,
Professor
Titular,
Disciplina
de
Endocrinologia,
Departamento de Medicina, Escola Paulista de Medicina, UNIFESP.
3. Mario Vaisman, Professor Titular, Disciplina de Endocrinologia, Departamento de
Clínica Médica, Universidade Federal do Rio de Janeiro (UFRJ).
4. João
Hamilton
Departamento
de
Romaldini,
Professor
Titular,
Disciplina
Medicina,
Pontifícia
Universidade
de
Católica
Endocrinologia,
de
Campinas
(PUCCAMP).
5. Hans Graf, Professor Associado, Disciplina de Endocrinologia, Departamento de
Clínica Médica, Universidade Federal do Paraná (UFPr).
Suplentes:
1. Osmar Monte, Professor Titular, Disciplina de Endocrinologia, Departamento de
Medicina, Faculdade de Ciências Médicas da Santa Casa de São Paulo.
2. João
Roberto
Maciel
Martins,
Médico
Assistente
Doutor,
Disciplina
de
Endocrinologia, Escola Paulista de Medicina, UNIFESP
4
À Luciana, Fábio e Caio
5
AGRADECIMENTOS
À comunidade Nipo-Brasileira de Bauru, pela disponibilidade, compreensão e
organização que possibilitaram a realização deste estudo.
Ao Grupo de Estudo Nipo-Brasileiro em Diabetes, especialmente ao Prof. Dr. Laércio
Franco, Profa. Dra. Amélia Hirai e Prof. Dra. Sueli Gimeno, pela possibilidade da
realização deste estudo e apoio.
À Teresa Kasamatsu, pela dedicação na realização das determinações laboratoriais,
por sua disponibilidade e proximidade, pelo zelo e apreço para com as minhas
pendências, pelo acolhimento e pela amizade.
À Prof. Dra. Sandra Roberta Ferreira, pela disponibilidade, cordialidade, sugestões e
revisão de manuscritos.
À Sirlei Siani, pelo extensivo trabalho de análise estatística e pela compreensão nos
momentos de discordância e tensão.
À Ângela Faria, pela assistência administrativa, acolhimento e cordialidade.
Ao Gilberto Koiti, Ilda Kunii e Patrícia Amioka, pelo auxílio nas determinações dos
anticorpos anti-peroxidase tiroidiana.
À Amarylis, pela assistência nos trâmites burocráticos da pós-graduação.
À Prof. Dra. Heloísa Villar, pelo auxílio na exaustiva avaliação clínica da população
Nipo-Brasileira de Bauru, incentivo e amizade.
Aos queridos Magnus, Cléber, João e Janete, pelo acolhimento, incentivo, sugestões e
pelo exemplo de dedicação, desprendimento e ética.
6
Aos meus preceptores do Hospital do Servidor Público em São Paulo, Chady, Mozart,
Rubens, Horácio, Nazareth e Marisa, com os quais aprendi a tiroidologia clínica.
Ao Prof. Dr. João Romaldini, meu mentor e amigo, pela presença constante orientando
minha vida acadêmica, profissional e pessoal, por me compreender e aos meus limites
e por ter me apresentado à tiroide com seu olhar.
Ao Rômulo, pela compreensão, tolerância e resignação e por não ter desistido de
tentar.
À minha irmã, Dalcira e aos queridos Ideval, Felipe e Tiago, pela proximidade,
cumplicidade e incentivo. Pelo exemplo de humildade, desprendimento e doação. Pela
referência, o norte, um porto a qualquer hora e por qualquer razão.
Aos meus pais, Alberto e Dalcira, pela educação calcada na simplicidade, na intuição,
no amor e no conceito de família, por todas as oportunidades que me foram oferecidas,
pelo exemplo do trabalho incansável e ético como meio de transformação social e por
tudo o que sou.
À Luciana, pela paz e o conforto do seu amor, pelo olhar que ilumina a minha vida, pela
alegria e bom humor contagiantes, pela amizade e cumplicidade, pela compreensão e
incentivo.
Aos meus filhos, Fábio e Caio, por tudo que me ensinaram, pela oportunidade dos
desafios, pelo amor infinito, por compreenderem os meus momentos de caserna e por
não me permitirem a reclusão.
7
AGRADECIMENTOS ESPECIAIS
À Prof. Dra. Luiza K. Matsumura, pela co-orientação desta tese, divisão de trabalho,
pelo papel facilitador e moderador, pelas palavras de incentivo, torcida e cumplicidade,
pelo ombro amigo nos momentos mais difíceis e pelo privilégio da amizade.
Ao Prof. Dr. Rui M. B. Maciel, pela orientação desta tese, pelos caminhos que foram se
abrindo, pelas oportunidades que extrapolaram os limites desta tese, pelos momentos
transformadores, pela inclusão, pela confiança que foi se conquistando, pela amizade
que foi se fazendo, pela seriedade do trabalho, pelo respeito, pela dignidade, pela
ética, pelo exemplo de tutor e de mentor.
8
“...Sê todo em cada coisa.
Põe quanto és no mínimo que fazes.
Assim em cada lago a lua toda.
Brilha, porque alta vive”
Ricardo Reis, 1933
9
ÍNDICE
Introdução .......................................................................................... 11
Objetivos ............................................................................................ 15
Apresentação ..................................................................................... 16
Comentários e Perspectivas............................................................... 18
Conclusões ......................................................................................... 22
Referências ........................................................................................ 24
Anexo 1 (Trabalho No. 01)................................................................... 32
Anexo 2 (Trabalho No. 02)................................................................... 42
Anexo 3 (Trabalho No.03).................................................................... 49
Anexo 4 (Trabalho No.04).....................................................................59
Anexo 5 (Trabalho No. 05)....................................................................70
10
INTRODUÇÃO
As disfunções da glândula tiroide são comuns e posicionam-se entre as
condições médicas de maior prevalência na população geral. No Reino Unido (1),
estima-se que aproximadamente 15% dos adultos tenham alguma disfunção tiroidiana,
mas concentrações elevadas do TSH podem atingir até 20% das mulheres idosas. Nos
Estados Unidos da América (2), mais de 10 milhões de pessoas sabiam serem
portadoras de alguma doença tiroidiana, enquanto outros nove milhões de não
suspeitos apresentaram evidências bioquímicas de alguma disfunção. Estudos
epidemiológicos (3-5) em outras regiões do mundo corroboram com esses achados,
mas há variações consideráveis, particularmente em função dos critérios de seleção
empregados, como idade, sexo e etnia, da ingestão de iodo na dieta e dos ensaios
hormonais utilizados (6).
No Brasil, há escassos estudos epidemiológicos e a prevalência das
disfunções tiroidianas não é bem conhecida. No Nordeste brasileiro (7), a freqüência de
alterações tiroidianas em uma amostra não representativa de 210 indivíduos do
município de Cabeceiras, na Paraíba, foi de 20%. Na região metropolitana de São
Paulo, a prevalência de doença autoimune da tiroide foi de 16,9% em uma amostra
representativa de 1085 indivíduos, sendo que o hipertiroidismo foi encontrado em 3,3%
e o hipotiroidismo em 8,0% deles (8). Na cidade do Rio de Janeiro (4,9), em uma
amostra representativa de 1220 mulheres, a prevalência de hipotiroidismo foi de 12,3%,
de hipertiroidismo de 3,7% e de autoimunidade tiroidiana de 13%. Mais recentemente,
a freqüência de doença tiroidiana subclínica foi investigada em 314 funcionárias da
Universidade de São Paulo (10). As taxas de hipotiroidismo subclínico e de
11
hipertiroidismo subclínico foram respectivamente de 7,3% e 5,1%, enquanto as taxa de
anticorpos antiperoxidase positivos foi de 16,2%.
Entre os distúrbios tiroidianos, as doenças autoimunes da tiroide e as
disfunções tiroidianas subclínicas são as de maior prevalência e também as que
despertam maior interesse de clínicos e pesquisadores.
As doenças autoimunes da tiroide afetam de 2 a 5% da população geral, em
especial mulheres adultas e idosas (11). São causadas possivelmente pela
combinação de múltiplos fatores, genéticos e ambientais, mas a identificação e o papel
de cada um desses fatores de suscetibilidade ainda não estão bem definidos (11-14).
Uma característica curiosa e intrigante das doenças autoimunes da tiroide é a sua forte
preponderância no sexo feminino (6,15). Possíveis explicações incluem os efeitos dos
hormônios sexuais no sistema imune (16), alterações no cromossomo X (17) e o
microquimerismo fetal (18-20), um conceito que envolve a transferência de células
fetais para a circulação materna, que após longo período de tempo em tecidos
maternos, participariam do desencadeamento da autoimunidade tiroidiana. Assim,
paridade poderia se constituir em um fator de risco para doenças autoimunes da tiroide,
explicando a preponderância feminina, mas há escassos estudos populacionais
explorando esta hipótese e os resultados são conflitantes (21-23).
Especula-se que a concentração de iodo na dieta exerça um papel de
principal modulador ambiental do processo de autoimunidade tiroidiana. Em geral, sua
deficiência atenua, enquanto o excesso de iodo acelera a indução de tiroidite
autoimune em indivíduos geneticamente suscetíveis (24). Desta forma, em regiões
onde a ingestão de iodo é elevada, como no Japão (3), a prevalência de autoimunidade
12
é maior quando comparada a regiões onde a ingestão de iodo é normal ou
relativamente baixa (25-26).
As disfunções tiroidianas subclínicas caracterizam-se pela presença de
concentrações séricas anormais do TSH em face de concentrações normais dos
hormônios tiroidianos (27). Afetam até 20% da população adulta, sendo o
hipotiroidismo subclínico mais comum que o hipertiroidismo subclínico, e ambos, em
geral, são mais prevalentes no sexo feminino e em idosos (1,2). Apesar da elevada
prevalência na comunidade e do aumento do diagnóstico na prática diária, o significado
clínico e a necessidade de tratamento dessas condições permanecem controversos
(28) e motivo de intensos debates (29-31). Na última década, um número crescente de
estudos clínicos (32-35), populacionais (36-44) e de meta-análise (45-47) procurou
explorar os efeitos das doenças tiroidianas subclínicas no sistema cardiovascular e na
expectativa de vida, mas os resultados são divergentes.
Nesta Tese, apresentamos os resultados publicados, até o momento, de um
estudo populacional sobre as disfunções tiroidianas na comunidade Nipo-Brasileira de
Bauru, cidade desenvolvida localizada na região centro-oeste do Estado de São Paulo,
distante 450 km da capital paulista. Esta população mostrou-se organizada e
cooperativa em relação à realização de inquéritos clínicos e epidemiológicos anteriores
(48) coordenados pelo “Grupo de Estudo Nipo-Brasileiro em Diabetes” do
Departamento de Medicina Preventiva da Universidade Federal de São Paulo. Entre as
características desta população, sua elevada prevalência de diabetes, obesidade,
dislipidemia e hipertensão arterial (49) serviriam ao nosso interesse de explorar
possíveis associações das disfunções tiroidianas subclínicas com fatores de risco
cardiovascular e mortalidade. Além disso, trata-se de uma população não miscigenada
que sofreu significativa aculturação no que se refere a hábitos dietéticos, aqui
13
interessando principalmente o conteúdo de iodo na dieta, visto que a ingestão de iodo
no interior de São Paulo é nitidamente menor em comparação à do Japão (50). Por
tratar-se de uma população étnica estável e geneticamente homogênea, a comparação
de parâmetros de autoimunidade tiroidiana com dados provenientes de estudos em
populações japonesas residentes no Japão poderia ser um modelo ideal para estudar o
papel de fatores ambientais de susceptibilidade, particularmente do conteúdo do iodo
na dieta, no desenvolvimento da autoimunidade tiroidiana.
Em 2000, após realização de censo demográfico, todos os indivíduos com
mais de 30 anos foram convidados (n = 1751) para participação no estudo, dos quais
1330 aceitaram. Na avaliação basal, todos os participantes responderam a
questionários padronizados sobre estilo de vida, hábitos alimentares e sociais,
antecedentes médico, pessoal e familiar, uso de medicamentos, paridade e tabagismo.
Um questionário específico sobre doenças tiroidianas, incluindo antecedentes pessoais
e familiares, sinais e sintomas clínicos, uso atual ou prévio de hormônios tiroidianos,
drogas antitiroidianas ou de drogas que pudessem interferir com a função tiroidiana
(glicocorticóides, amiodarona e lítio), foi aplicado por três tiroidologistas experientes,
que também realizaram o exame físico geral e da tiroide em todos os indivíduos da
população.
14
OBJETIVOS
1. Avaliar a prevalência e o espectro das disfunções tiroidianas na comunidade NipoBrasileira de Bauru.
2. Investigar o papel de possíveis fatores ambientais, como o conteúdo de iodo na dieta,
no espectro das disfunções tiroidianas.
3. Investigar o papel de fatores biológicos, como paridade e o microquimerismo fetal, na
prevalência de autoimunidade tiroidiana.
4. Explorar potenciais associações das disfunções tiroidianas subclínicas com o risco
cardiovascular e mortalidade.
15
APRESENTAÇÃO
Nesta tese de doutorado apresentamos 5 trabalhos completos sobre aspectos
epidemiológicos das disfunções tiroidianas na população Nipo-Brasileira de Bauru, em
uma linha de pesquisa focada particularmente nas disfunções tiroidianas subclínicas e
nas doenças autoimunes da tiroide. A pesquisa foi financiada pela Fundação de Apoio
à Pesquisa no Estado de São Paulo, FAPESP (auxílio à pesquisa 06/59737-9,
“Epidemiologia das disfunções tiroidianas na população Nipo-Brasileira de Bauru).
Os trabalhos são:
1. “Patogênese das doenças autoimunes da tiroide”, de autoria de José A. Sgarbi e Rui
M. B. Maciel, publicado na revista “Arquivos Brasileiros de Endocrinologia e
Metabologia” (Arq Bras Endocrinol Metab 2009; 53: 5-14) - Anexo 1.
2. “Parity is not related to autoimmune thyroid disease in a population-based study of
Japanese-Brazilians”, de autoria de José A. Sgarbi, Teresa S. Kasamatsu, Luiza K.
Matsumura e Rui M.B. Maciel, publicado na revista “Thyroid” (Thyroid 2010; 20: 1151 1156) – Anexo 2.
3. “Subclinical thyroid dysfunctions are independent risk factors for mortality in a 7.5year follow-up: the Japanese-Brazilian thyroid study”, de autoria de José A. Sgarbi,
Luiza K. Matsumura, Teresa S. Kasamatsu, Sandra R. Ferreira e Rui M. B. Maciel,
publicado na revista “European Journal of Endocrinology” (Eur J Endocrinol. 2010; 162:
569 - 577) – Anexo 3.
4. “Subclinical hypothyroidism and the risk of coronary heart disease and mortality”, de
autoria de Nicolas Rodondi, Wendy P. J. den Elzen, Douglas C. Bauer, Anne R.
16
Cappola, Salman Razvi, John P. Walsh, Bjørn O. Åsvold, Giorgio Iervasi, Misa
Imaizumi, Team H. Collet, Alexandra Bremner, Patrick Maisonneuve, José A. Sgarbi,
Khaw KT, Mark Vanderpump, Anne B. Newman, Jacques Cornuz, Jayne A. Franklyn,
Westendorp RG, Eric Vittinghoff, Jacobijn Gussekloo; for the Thyroid Studies
Collaboration, publicado na revista
“Journal of the American Medical Association”
(JAMA 2010; 304:1365-1374) – Anexo 4.
5. “Associations between subclinical hypothyroidism and traditional and nontraditional
cardiovascular risk factors”, de autoria de José A. Sgarbi, Luiza K. Matsumura, Teresa
S. Kasamatsu, Heloisa H. Villar, Sandra R. Ferreira e Rui M. B. Maciel, Manuscristo em
preparo para publicação – Anexo 5.
17
COMENTÁRIOS E PERSPECTIVAS
A presente tese apresenta quatro trabalhos publicados e um submetido para
publicação em periódicos expressivos da literatura médica nacional e internacional. O
primeiro trabalho (11), foi uma revisão sobre a patogênese das doenças autoimunes da
tiroide, com foco nos fatores genéticos e ambientais de susceptibilidade, publicado na
revista “Arquivos Brasileiros de Endocrinologia e Metabologia” (Anexo 1). Discorremos
sobre o papel de fatores ambientais, particularmente do conteúdo do iodo na dieta, e
de fatores biológicos, como a paridade, no desenvolvimento das doenças autoimunes
da tiroide. Pretendíamos fazer uma introdução para publicações posteriores que
comparariam a prevalência de doenças autoimunes entre a população japonesa
residente no Japão e uma população geneticamente similar residente no Brasil. Tratase de um modelo de estudo ideal para demonstrar o impacto da aculturação de uma
população e diferenças ambientais no desenvolvimento de doenças autoimunes, neste
caso importando as diferenças conhecidas do conteúdo do iodo na dieta entre as duas
populações. Os resultados esperados seriam de uma maior prevalência de
autoimunidade tiroidiana entre os japoneses residentes no Japão em razão da elevada
ingestão de iodo, mas dados preliminares de toda a população de migrantes japoneses
residentes em Bauru (ainda não publicados) e especificamente da população feminina,
objeto de nossa segunda publicação (50), na revista “Thyroid” (Anexo 2), não
confirmaram tal hipótese. Nesta mesma publicação, enfatizamos que paridade não foi
um fator biológico de relevância no desenvolvimento de autoimunidade tiroidiana na
população feminina estudada. Este dado corrobora com outros dois estudos
populacionais (21,22), trazendo fortes argumentos epidemiológicos contra o papel do
microquimerismo fetal como modulador ou desencadeador da autoimunidade tiroidiana,
18
apontado como um possível mecanismo para explicar a intrigante preponderância
feminina (1-3, 6,15) nas doenças autoimunes da tiroide.
O terceiro estudo apresentado nesta tese (Anexo 3), publicado na revista
“European Journal of Endocrinology” (44), traz grande contribuição ao entendimento do
significado clínico das doenças tiroidianas subclínicas e sobre a influência dessas
disfunções na expectativa de vida. A publicação, que mereceu comentários especiais
em editorial da revista (51), confirmou a elevada prevalência das disfunções tiroidianas
subclínicas na população geral e mostrou associação significativa do hipertiroidismo
subclínico com a mortalidade geral e cardiovascular, e do hipotiroidismo subclínico com
a mortalidade geral. O impacto desta publicação foi imediato, de tal forma que o estudo
foi incluído em uma revisão sistemática da literatura de estudos publicados desde o
ano de 1950 até maio de 2010, sobre a associação entre doenças tiroidianas
subclínicas e mortalidade. Entre as 4440 publicações identificadas, apenas 11 estudos
prospectivos
contemplaram
os
critérios
rigorosos
de
seleção
(eFigure
at
HTTP://www.jama.com). Para confirmar nossos resultados e a qualidade dos dados,
abrimos nossa base de dados para uma análise independente, o que resultou em
convite para participarmos como membros ativos de um grupo internacional de
pesquisadores, clínicos e epidemiologistas, envolvidos em estudos populacionais sobre
as disfunções tiroidianas subclínicas, o que denominou-se “The Thyroid Studies
Collaboration Group”. A participação neste grupo resultou em publicação no “JAMA”
(52), periódico de elevado impacto e reconhecida credibilidade, do quarto estudo
apresentado nesta tese (Anexo 4). O estudo demonstra evidência robusta da
associação do hipotiroidismo subclínico com um elevado risco de eventos de doença
coronariana e com mortalidade por doença coronariana.
19
No quinto estudo desta tese (Anexo 5), apresentamos um manuscrito em
preparo para publicação que explora possíveis associações entre o hipotiroidismo
subclínico e fatores de risco cardiovascular, tradicionais e não tradicionais. Trata-se de
estudo de grande interesse, pois se de um lado nos parece estabelecida a associação
do
hipotiroidismo
subclínico
com
a
mortalidade,
particularmente
por
causa
cardiovascular, por outro lado, os mecanismos permanecem ainda não conhecidos.
Este estudo contribui com a literatura no sentido de não encontrar associação entre o
hipotiroidismo subclínico com fatores de risco cardiovascular, clássicos e não clássicos,
sugerindo que outros mecanismos, que não aqueles associados ao processo
inflamatório da aterosclerose, devam estar envolvidos no mecanismo de morte de
pacientes com hipotiroidismo subclínico.
Desta forma, acreditamos ter contribuído de modo significativo com a
literatura médica, particularmente no que se refere ao significado clínico das disfunções
tiroidianas subclínicas, inclusive com possibilidade de mudanças dos paradigmas no
tratamento dessas disfunções. Novas contribuições resultantes deste estudo
epidemiológico serão brevemente submetidas para publicação, em resposta aos
assuntos, ainda passíveis de controvérsia e pouco explorados na literatura. De nosso
conhecimento, apenas um estudo epidemiológico prévio (53) investigou a presença de
sinais e sintomas tiroidianos em indivíduos com disfunção tiroidiana subclínica na
população geral. Esses dados são de grande relevância, pois indivíduos assintomáticos
de uma população poderiam permanecer com uma disfunção tiroidiana subclínica por
longo período de tempo, aumentando sua exposição aos efeitos deletérios da doença,
o risco de complicações e de morte. Além disso, a análise de novos dados, obtidos em
reavaliação mais recente de parte desta população, permitirá estudar a história natural
20
das disfunções tiroidianas subclínicas e das doenças autoimunes tiroidianas nesta
população. Novos estudos em colaboração, resultantes da nossa participação no
“Thyroid Studies Collaboration Group” já estão em fase de análise final e confecção de
manuscrito, sendo o próximo um estudo de análise de dados individuais sobre
associações entre o hipertiroidismo subclínico e mortalidade.
Nosso próximo desafio, entretanto, será desenvolver e conduzir um projeto
temático colaborativo, de estudo epidemiológico que estude o comportamento e as
variadas facetas das doenças tiroidianas na população brasileira, importando
particularmente as doenças autoimunes da tiroide, as disfunções tiroidianas subclínicas
e o câncer de tiroide.
Nosso
compromisso
será
com
o
fortalecimento
da
Disciplina
de
Endocrinologia e Metabologia da Faculdade de Medicina de Marília, tornando-a um
centro formador de novos pesquisadores em tiroide.
21
CONCLUSÕES
1. A prevalência de hipotiroidismo subclínico na população Nipo-Brasileira de Bauru é
semelhante à de outras populações ocidentais suficientes em iodo, enquanto a
prevalência de hipertiroidismo subclínico mostrou-se elevada (Estudo No. 03).
2. A concentração mediana de iodo urinário na população Nipo-Brasileira de Bauru foi de
210 µg/L, significando que esta população é suficiente em iodo (Estudo No. 03).
Nenhuma associação foi encontrada entre o conteúdo de iodo na dieta e a o espectro
das disfunções tiroidianas nesta população (Estudos No. 02 e No. 03).
3. Paridade não foi um fator de risco para doença autoimune da tiroide em NipoBrasileiras. Os dados não favorecem a hipótese do microquimerismo fetal no
desenvolvimento da autoimunidade tiroidiana (Estudo No. 02). Aborto e uso de
estrógeno também não se associaram com autoimunidade tiroidiana nesta população
(Estudo No. 02).
4. Doença tiroidiana subclínica não se associou com maior prevalência de doença
cardiovascular
(Estudo
No.
03).
Hipertiroidismo
subclínico
associou-se
significativamente com maior risco de mortalidade total e por causa cardiovascular,
enquanto que o hipotiroidismo subclínico associou-se apenas com maior risco de
mortalidade total (Estudo No. 03).
5. Hipotiroidismo subclínico foi um fator de risco significativo para desenvolvimento de
doença arterial coronariana e de morte por doença arterial coronariana (Estudo No. 04).
Níveis séricos do TSH ≥ 7,0 mU/L associou-se com maior risco de morte por doença
arterial coronariana (Estudo No. 04).
22
6. Hipotiroidismo subclínico não se associou com síndrome metabólica ou com fatores de
risco cardiovascular, clássicos e não clássicos, na população Nipo-Brasileira, exceto
com hipertrigliceridemia no sexo feminino (Estudo No. 05).
23
REFERÊNCIAS
1. Tunbridge WM, Evered DC, Hall R, Appleton D, Brewis M, Clark F, Evans JG,
Young E, Bird T, Smith PA. The spectrum of thyroid disease in a community:
the Whickham survey. Clin Endocrinol (Oxf) 1977; 7: 481- 493.
2. Hollowell JG, Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer
CA, Braverman LE. Serum TSH, T4, and thyroid antibodies in the United
States population (1988 to 1994): National Health and Nutrition Examination
Survey (NHANES III). J Clin Endocrinol Metab 2002; 87: 489 - 499.
3. Konno N, Yuri K, Taguchi H, Miura K, Taguchi S, Hagiwara K, Murakami S.
Screening for thyroid diseases in an iodine sufficient area with sensitive
thyrotrophin assays, and serum thyroid autoantibody and urinary iodine
determinations. Clin Endocrinol 1993; 38: 273-281.
4. Sichieri R, Baima J, Henriques J, Vasconcellos M, Marante T, Kumagai S,
Vaisman M. Prevalence of thyroid disease and positive antitiroperoxidase
among 1,500 women 35 year old and older: a population-based survey in the
city of Rio de Janeiro, Brazil. Thyroid 2005; 15 (Supp 1): S-42 abs 118.
5. Walsh JP, Bremner AP, Bulsara MK, O'Leary P, Leedman PJ, Feddema P,
Michelangeli V. Subclinical thyroid dysfunction as a risk factor for
cardiovascular disease. Arch Int Med 2005; 165: 2467-2472.
6. Wang C, Crapo LM. The epidemiology of thyroid disease and implications for
screening. Endocrinol Metab North Am 1997, 26: 189-218.
7. Pontes AN, Adan LF, Costa AM, Benício AV, Silva, CRA, Morais, RM,
Pedrosa VC. Prevalência de Doenças da Tireóide em Uma Comunidade do
24
Nordeste Brasileiro. Arq Bras Endocrinol Metab 2002; 46: 544-549.
8. Camargo RY, Tomimori EK, Neves SC, Rubio IG, Galrão AL, Knobel M,
Medeiros-Neto G. Thyroid and the environment: exposure to excessive
nutritional iodine increases the prevalence of thyroid disorders in Sao Paulo,
Brazil. Eur J Endocrinol 2008; 159: 293-299.
9.
Sichieri R, Baima J, Marante T, de Vasconcellos MT, Moura AS, Vaisman M.
Low prevalence of hypothyroidism among black and Mulatto people in a
population-based study of Brazilian women. Clin Endocrinol (Oxf) 2007; 66:
803-807.
10. Diaz-Olmos R, Nogueira AC, Penalva DQ, Lotufo PA, Benseñor IM.
Frequency of subclinical thyroid dysfunction and risk factors for cardiovascular
disease among women at a workplace. Sao Paulo Med J. 2010; 128: 18-23.
11. Sgarbi JA, Maciel RMB. Pathogenesis of autoimmune thyroid diseases. Arq
Bras Endocrinol Metab 2009, 53: 5-14.
12. Prummel MF, Strieder T, Wiersinga WM. The environment and autoimmune
thyroid disease. Eur J Endocrinol 2004; 150: 605-618.
13. Huber A, Tomer Y. The etiology of autoimmune thyroid disease: a story of
genes and environment. J Autoimmun 2009, 32: 231-239.
14. Burek CL, Talor MV. Environmental triggers of autoimmune thyroiditis. J
Autoimmun 2009, 33: 183-189.
15. Tumbridge WM, Vanderpump MPJ. Population screening for autoimmune
thyroid disease. Endocrinol Metab Clin North Am 2000; 29: 239-253.
16. McCombe PA, Greer JM, Mackay IR. Sexual dimorphism in autoimmune
25
disease. Curr Mol Med 2009, 9: 1058-1079.
17. Brix TH, Knudsen GP, Kristiansen M, Kyvik KO, Orstavik KH, Hegedus L. High
frequency of skewed X-chromosome inactivation in females with autoimmune
thyroid disease: a possible explanation for the female predisposition to thyroid
autoimmunity. J Clin Endocrinol Metab 2005, 90: 5949-5953.
18. Renné C, Ramos Lopez E, Steimle-Grauer SA, Ziolkowski P, Pani MA, Luther
C, Holzer K, Encke A, Wahl RA, Bechstein WO, Usadel KH, Hansmann ML,
Badenhoop K. Thyroid fetal male microchimerism in mothers with thyroid
disorders: presence of y-choromosomal immunofluorescence in thyroidinfiltrating lymphocytes is more prevalent in Hashimoto’s thyroiditis and
Graves’ disease than in follicular adenomas. J Clin Endocrinol Metab 2004;
89: 5810-5814.
19. Klintschar M, Immel UD, Kehlen A, Schwaiger P, Mustafa T, Mannweiler S,
Regauer S, Kleiber M, Hoang-Vu C. Fetal microchimerism in Hashimoto’s
thyroiditis: a quantitative approach. Eur J Endocrinol 2006; 154: 237-241.
20. Brix TH, Hansen PS, Kyvik KO, Hagedus L. Aggregation of thyroid
autoantibodies in twins from opposite-sex pairs suggests that microchimerism
may play a role in the early stages of thyroid autoimmunity. J Clin Endocrinol
Metab 2009; 94: 4439-4443.
21. Walsh JP, Bremner AP, Bulsara MK, O'Leary P, Leedman PJ, Feddema P,
Michelangeli V. Parity and the risk of autoimmune thyroid disease: a
community-based study. J Clin Endocrinol Metab 2005; 90: 5309-5312.
22. Bülow Pedersen I, Laurberg P, Knudsen N, Jørgensen T, Perrild H, Ovesen L,
Rasmussen LB. Lack of association between thyroid autoantibodies and parity
26
in a population study argues against microchimerism as a trigger of thyroid
autoimmunity. Eur J Endocrinol 2006;154:39-45.
23. Friedrich N, Schwarz S, Thonack J, Jonh U, Wallaschofski H, Volzke H.
Association between parity and autoimmune thyroiditis in a general female
population. Autoimmunity 2008; 41: 174-180.
24. Laurberg P, Pedersen B, Knudsen N, Ovesen L, Andersen S. Environmental
iodine intake affects the type of nonmalignant thyroid disease. Thyroid 2001;
111: 457-69.
25. Zois C, Stavrou I, Svarna E, Seferiadis K, Tsatsoulis A. Natural course of
autoimmune thyroiditis after elimination of iodine deficiency in northwestern
Greece. Thyroid 2006; 16: 289-293.
26. Teng W, Shan Z, Teng X, Guan H, Li Y, Teng D, Jin Y, Yu X, Fan C, Chong
W, Yang F, Dai H, Yu Y, Li J, Chen Y, Zhao D, Shi X, Hu F, Mao J, Gu X,
Yang R, Tong Y, Wang W, Gao T, Li C. Effect of iodine intake on thyroid
diseases in China. N Engl J Med 2006; 354: 2783-2793.
27. Romaldini JH, Sgarbi JA, Farah CS. Subclinical thyroid disease: subclinical
hypothyroidism and hyperthyroidism. Arq Bras Endocrinol Metab 2004;
48:147-58.
28. Biondi B, Cooper DS. The clinical significance of subclinical thyroid
dysfunction. Endocrine Review 2008; 29: 76-131.
29. Helfand M. Screening for subclinical thyroid dysfunction in nonpregnant adults:
a summary of the evidence for the U.S. Preventive Service Task Force. Ann
Intern Med 2004; 140:128 – 141
27
30. Surks MI, Ortiz E, Daniels G et al. Subclinical thyroid disease: scientific review
and guidelines for diagnosis and management. JAMA 2004; 291: 228-238.
31. Gharib H, Tuttle M, Baskin HJ, Fish LH, Singer PA, McDermott MT. Subclinical
thyroid dysfunction: a joint statement on management from the American
Association of Clinical Endocrinologists, the American Thyroid Association,
and The Endocrine Society. J Clin Endocrinol Metab 2005; 90: 581-585.
32. Biondi B, Palmieri EA, Fazio S et al. Endogenous subclinical hyperthyroidism
affects quality of life and cardiac morphology and function in young and
middle-aged patients. J Clin Endocrinol Metab. 2000; 85: 4701–4705.
33. Sgarbi JA, Villaça F, Garbeline B et al. The effects of early antithyroid therapy
for
endogenous
subclinical
hyperthyroidism
on
clinical
and
heart
abnormalities. J Clin Endocrinol Metab 2003; 88: 1672-1677.
34. Monzani F, Di Bello V, Caraccio N et al. Effect of levothyroxine on cardiac
function and structure in subclinical hypothyroidism: a double blind, placebocontrolled study. J Clin Endocrinol Metab 2001; 86:1110-1115.
35. Vitale G, Galderisi M, Lupoli GA et al. Left ventricular myocardial impairment in
subclinical hypothyroidism assessed by a new ultrasound tool: pulsed tissue
Doppler. J Clin Endocrinol Metab 2002; 87: 4350-4355
36. Hak AE, Pols HA, Visser TJ, Drexhage HA, Hofman A, Witteman JC.
Subclinical hypothyroidism is an independent risk factor for atherosclerosis
and myocardial infarction in elderly women: the Rotterdam Study. Ann Intern
Med 2000; 132: 270-278.
37. Imaizumi M, Akahoshi M, Ichimaru S, et al. Risk for ischemic heart disease
28
and all-cause mortality in subclinical hypothyroidism. J Clin Endocrinol Metab
2004; 89:3365–3370.
38. Cappola AR, Fried LP, Arnold AM et al. Thyroid status, cardiovascular risk,
and mortality in older adults. JAMA 2006; 295: 1033-1041.
39. Gammage MD, Parle JV, Holder HL et al. Association between serum free
thyroxine concentration and atrial fibrillation. Arch Intern Med 2007; 167: 928934.
40. Parle JV, Maisonneuve P, Sheppard MC, Boyle P, Franklyn JA. Prediction of
all-cause and cardiovascular mortality in elderly people from one low serum
thyrotropin result: a 10-year cohort study. Lancet 2001; 358: 861-865.
41. Gussekloo J, van Exel E, de Craen AJ, Meinders AE, Frolich M, Westendorp
RG. Thyroid status, disability and cognitive function, and survival in old age.
JAMA 2004; 292: 2591–2599.
42. Rodondi N, Newman AB, Vittinghoff E et al. Subclinical hypothyroidism and
the risk of heart failure, other cardiovascular events, and death. Arch Intern
Med. 2005; 165: 2460-2466.
43. van den Beld AW, Visser TJ, Feelders RA, Grobbee DE, Lamberts SW.
Thyroid hormone concentrations, disease, physical function, and mortality in
elderly men. J Clin Endocrinol Metab 2005; 90: 6403–6409.
44. Sgarbi JA, Matsumura LK, Kasamatsu TS, Ferreira SR, Maciel RM.
Subclinical thyroid dysfunctions are independent risk factors for mortality in a
7.5-year follow-up: the Japanese-Brazilian thyroid study. Eur J Endocrinol
2010; 162: 569-77.
29
45. Ochs N, Auer R, Bauer DC, Nanchen D, Gussekloo J, Cornuz J, Rodondi N.
Meta-analysis: subclinical thyroid dysfunction and the risk for coronary heart
disease and mortality. Ann Int Med 2008; 148: 832-845.
46. Razvi S, Shakoor A, Vanderpump M, Weaver JU, Pearce SH The influence of
age on the relationship between subclinical hypothyroidism and ischemic heart
disease: a metaanalysis. J Clin Endocrinol Metab 2008; 93: 2998-3007.
47. Volzke H, Schwahn C, Wallaschofski H, Dorr M. Review: The association of
thyroid dysfunction with all-cause and circulatory mortality: is there a causal
relationship? J Clin Endocrinol Metab 2007; 92: 2421-2429.
48. Gimeno SG, Ferreira SR, Franco LJ, Hirai AT, Matsumura L, Moisés RS
Prevalence and 7-year incidence of Type II diabetes mellitus in a JapaneseBrazilian population: an alarming public health problem. Diabetologia 2002;
45: 1635-1638.
49. Rosenbaum P, Gimeno SG, Sanudo A, Franco LJ, Ferreira SR; JapaneseBrazilian Diabetes Study Group Analysis of criteria for metabolic syndrome in
a population-based study of Japanese-Brazilians. Diabetes Obes Metab 2005;
7: 352 – 359.
50. Esteves RZ, Kasamatsu TS, Kunii IS, Furuzawa GK, Vieira JGH, Maciel RMB.
Development of a semi-automated method for measuring urinary iodine and its
application in epidemiological studies in Brazilian school children. Arq Bras
Endocrinol Metab 2000; 51: 1477-1484.
51. Sgarbi JA, Kasamatsu TS, Matsumura LK, Maciel RM. Parity is not related to
autoimmune thyroid disease in a population-based study of JapaneseBrazilians. Thyroid 2010; 20: 1151-1156.
30
52. Biondi B. Invited commentary: Cardiovascular mortality in subclinical
hyperthyroidism: an ongoing dilemma. Eur J Endocrinol 2010; 162: 587-589.
53. Rodondi N, den Elzen WP, Bauer DC, Cappola AR, Razvi S, Walsh JP,
Asvold BO, Iervasi G, Imaizumi M, Collet TH, Bremner A, Maisonneuve P,
Sgarbi JA, Khaw KT, Vanderpump MP, Newman AB, Cornuz J, Franklyn JA,
Westendorp RG, Vittinghoff E, Gussekloo J; Thyroid Studies Collaboration.
Subclinical hypothyroidism and the risk of coronary heart disease and
mortality. JAMA 2010; 304: 1365-1374.
54. Canaris GJ, Manowitz NR, Mayor G, Ridgway EC The Colorado thyroid
disease prevalence study. Arch Intern Med 2000; 160: 526 - 534.
31
ANEXO 1
Trabalho No. 01
Patogênese das doenças autoimunes da tiroide
José A. Sgarbi e Rui M. B. Maciel
“Arquivos Brasileiros de Endocrinologia e Metabologia”
(Arq Bras Endocrinol Metab 2009; 53: 5-14)
32
revisão
Patogênese das doenças
tiroidianas autoimunes
Pathogenesis of autoimmune thyroid diseases
José Augusto Sgarbi1,2, Rui M. B. Maciel2
Resumo
A doença tiroidiana autoimune (DAIT), que afeta de 2% a 5% da população ocidental, é o transtorno autoimune órgão-específico mais comum. Sua apresentação clínica varia do hipertiroidismo da doença de Graves (DG) ao hipotiroidismo associado à tiroidite de Hashimoto (TH).
A exata etiologia da DAIT permanece desconhecida, mas a interação entre suscetibilidade
genética e fatores ambientais desencadeadores parece ser de fundamental importância no
seu desenvolvimento. Postula-se que fatores genéticos responderiam por 79% da suscetibilidade à DAIT e os ambientais por 21%. Genes imunomoduladores, como o complexo maior
de histocompatibilidade (MHC), antígeno-4 associado ao linfócito T citotóxico (CTLA-4), a
molécula CD40 e a proteína tirosina fosfatase-22 (PTPN22) e os genes específicos da glândula tiróide, como receptor do TSH (TSHR) e tiroglobulina (TG) têm sido identificados. A
natureza exata do envolvimento do meio ambiente no desenvolvimento da DAIT não é bem
conhecida, mas vários fatores ambientais têm sido envolvidos, como o conteúdo de iodo na
dieta, estresse, drogas e infecções. Entretanto, não há evidência clara de causalidade e os
mecanismos pelos quais fatores ambientais desencadeariam a autoimunidade tiroidiana,
em indivíduos geneticamente predispostos, ainda permanecem não completamente entendidos. O conhecimento dos mecanismos precisos de interação entre fatores ambientais e
genes na indução da autoimunidade tiroidiana poderia resultar desenvolvimento de novas
estratégias de prevenção e tratamento. Arq Bras Endocrinol Metab. 2009;53(1):5-14.
1
Disciplina de Endocrinologia,
Faculdade de Medicina de Marília
(Famema), Marília, SP, Brasil
2
Disciplina de Endocrinologia,
Departamento de Medicina,
Escola Paulista de Medicina,
Universidade Federal de São Paulo
(EPM-Unifesp); São Paulo, SP, Brasil
Descritores
Tiroidite autoimune; doença de Graves; hipotiroidismo autoimune; genes de suscetibilidade
Autoimmune thyroid disease (AITD) is the most common organ-specific autoimmune disorder affecting 2% to 5% of the population in Western countries. Clinical presentation varies from hyperthyroidism in Graves’ Disease to hypothyroidism in Hashimoto’s thyroiditis.
While the exact etiology of thyroid autoimmunity is not known, interaction between genetic
susceptibility and environmental factors appears to be of fundamental importance to initiate
the process of thyroid autoimmunity. It has been postulated that 79% of the susceptibility
to develop AITD is attributed to genetic factors, while environmental factors contribute to
21%. The identified AITD susceptibility genes include immune-modulating genes, such as
the major histocompatibility complex (MHC), cytotoxic T lymphocyte antigen-4 (CTLA-4),
CD40 molecule, and protein tyrosine phosphatase-22 (PTPN22), and thyroid-specific genes,
including TSH receptor (TSHR) and thyroglobulin (TG). The exact nature of the role environmental factors play in AITD is still not well known, but the involvement of several factors
such as iodine diet content, stress, drugs and infections has been reported. However, there
is no clear evidence of causality and the mechanisms by which environmental factors trigger
thyroid autoimmunity in genetically predisposed individuals remain not fully understood.
Knowledge of the precise mechanisms of interaction between environmental factors and
genes in inducing thyroid autoimmunity could result in the development of new strategies
for prevention and treatment. Arq Bras Endocrinol Metab. 2009;53(1):5-14.
Endereço para correspondência:
Rui M. B. Maciel
Lab.Endocrinologia Molecular
Disciplina de Endocrinologia,
Dpto Medicina Unifesp
Rua Pedro de Toledo, 781, 12º andar
04029-032 São Paulo, SP
[email protected]
Recebido em 30/Mar/2008
Aceito em 7/Jul/2008
Copyright© ABE&M todos os direitos reservados.
Abstract
Keywords
Hashimoto’s thyroiditis; Graves disease; auto-immune hypothyroidism; susceptibility genes
Arq Bras Endocrinol Metab. 2009;53/1
5
Patogênese das doenças tiroidianas autoimunes
INTRODUÇÃO
Copyright© ABE&M todos os direitos reservados.
A
s doenças tiroidianas autoimunes (DAIT), consideradas como arquétipo das doenças autoimunes
órgão-específicas (1), afetam de 2% a 5% da população
geral, em especial mulheres adultas e idosos (2-3) e são
determinadas pela perda da autotolerância imunológica. São causadas possivelmente pela combinação de
múltiplos fatores, genéticos e ambientais, mas a identificação e o papel de cada um desses fatores de suscetibilidade ainda não estão bem definidos (4,5). Envolve
espectro de fenótipos cujos principais representantes
são a doença de Graves (DG) e a tiroidite de Hashimoto (TH), ambas caracterizadas pela presença de infiltrado linfocítico de intensidade variável e produção
de autoanticorpos tiroidianos dirigidos a antígenos específicos, determinantes da expressão clínica da enfermidade, que pode variar do hiper ao hipotiroidismo.
Outras formas de DAIT incluem a tiroidite pós-parto
(6), a tiroidite silenciosa (7), a tiroidite induzida por
α-interferon (8) e a tiroidite que acompanha as síndromes autoimunes poliglandulares (9). A tiroglobulina
(TG), a tiroperoxidase tiroidiana (TPO) e o receptor
do TSH (TSHR) são considerados os principais autoantígenos tiroidianos específicos na resposta autoimune tiroidiana (4,10-12).
Existem evidências sólidas da interação de múltiplos
fatores, genéticos e ambientais, para o desenvolvimento
da DAIT. A taxa de concordância para a doença em gêmeos homozigóticos, muito maior do que a encontrada
em gêmeos heterozigóticos e, ao mesmo tempo, a concordância menor que 100% em gêmeos homozigóticos,
implicam, respectivamente, a existência de fatores genéticos (13) e ambientais (1) no seu desenvolvimento.
Além disso, o fato de que imigrantes de países com baixa incidência de doença autoimune se adaptem à taxa de
incidência do novo país (1) fortalece a hipótese de que
suscetibilidade genética, em combinação com fatores
ambientais desencadeadores, iniciaria a resposta imune
aos antígenos tiroidianos.
A predisposição genética é provavelmente predominante, responsável por, aproximadamente, 80% da
suscetibilidade à DAIT (14), em que alelos do complexo maior de histocompatibilidade (MHC) e lócus
não-MHC, como polimorfismo no gene antígeno-4
associado ao linfócito T citotóxico (CTLA-4), têm sido
identificados como marcadores de suscetibilidade (15).
Por outro lado, pelo menos 20% da suscetibilidade seria determinada por fatores ambientais (14,16), como
tabagismo, estresse, infecção, selênio, iodo e drogas,
entre outros (1).
6
Os mecanismos pelos quais fatores ambientais desencadeariam resposta autoimune tiroidiana, em indivíduos
geneticamente suscetíveis, são ainda obscuros, mas a
interação entre gene e ambiente tem sido considerada
como processo fundamental para o desenvolvimento
da DAIT. Com esta perspectiva, foram revisadas, neste
artigo, evidências atuais da contribuição genética e ambiental na indução da autoimunidade tiroidiana.
Determinantes genéticos de suscetibilidade
A importância do envolvimento de fatores genéticos na
suscetibilidade para DAIT tem sido demonstrada por
modelos animais de desenvolvimento, pelo risco aumentado de DAIT em irmãos de indivíduos afetados e
pela maior taxa de concordância em gêmeos monozigóticos comparados a heterozigóticos (17).
A ocorrência familiar de DAIT tem sido reportada em
diversos estudos. Hall e Stanbury (18) mostraram que
33% dos irmãos de pacientes com DG ou TH desenvolveram DAIT e que 56% deles tinham autoanticorpos antitiroidianos. Além disso, apontaram que, em quase todos os
casos, pelo menos um dos pais do indivíduo afetado tinha
autoanticorpos tiroidianos, sugerindo que a herança para
a presença destes seria por traço dominante.
As comparações das taxas de concordância entre gêmeos homo e heterozigóticos tornaram-se evidências
robustas e irrefutáveis da influência genética na patogênese da DAIT. Estimou-se, por modelo biométrico,
que a hereditariedade seria responsável por 79% do desenvolvimento da DG (14), por 61% em homens e 72%
em mulheres do desenvolvimento da positividade do
anticorpo antiperoxidase tiroidiana (aTPO) e por 39%
em homens e 75% em mulheres do desenvolvimento do
anticorpo antitiroglobulina (aTG) (16). Assim, fatores
ambientais específicos explicariam o restante da suscetibilidade, significando que nenhum dos componentes
(genético e ambiental) poderia, isoladamente, determinar o desenvolvimento da DAIT.
A identificação de genes suscetíveis para DAIT tem sido
investigada, em geral, por meio de estudos populacionais
caso-controle, comparando-se a frequência de determinado alelo entre as populações doente e sadia. Recentes avanços no entendimento da base genética da DAIT revelaram
a participação de genes imunomoduladores e de genes
específicos da tiróide como os principais candidatos, mas,
por enquanto, a maioria dos achados é de polimorfismos
simples de um único nucleotídeo (SNPs) ou microssatélites (19). A Tabela 1 mostra a localização e o papel atribuído na suscetibilidade para DAIT desses genes.
Arq Bras Endocrinol Metab. 2009;53/1
Patogênese das doenças tiroidianas autoimunes
Tabela 1. Principais genes candidatos para suscetibilidade de DAIT.
Gene
Localização
Principais funções
Consequência funcional da variação
Fenótipo associado
MHC
6p21
Modula a ligação e a
apresentação de antígenos.
Alteração na ligação e apresentação do autoantígeno.
Expressão aberrante do MHC II.
DG
CTLA-4
2q33
Inibidor da ativação dos linfócitos T.
Hiperativação dos linfócitos T.
DG, TH
CD40
20q11
Ativação e proliferação dos linfócitos B.
Hiperativação dos linfócitos B.
DG
PTPN22
1p13
Inibidor da ativação dos linfócitos T.
Hiperativação dos linfócitos T.
DG, TH
8q24
Codifica a tiroglobulina.
Redução da imunotolerância à TG.
DG, TH
14q31
Codifica o TSHR.
Redução da imunotolerância ao TSHR.
DG
Genes imunomoduladores
Genes específicos
TG
TSHR
Os genes imunomoduladores de suscetibilidade
à DAIT, identificados e confirmados, são o MHC, o
CTLA-4, a molécula CD40 associada ao linfócito B e
a proteína tirosina fosfatase-22 (PTPN22) (15,19-22).
Outros genes, como o gene A relacionado à cadeia
MHC de classe I (MICA), o gene regulador da autoimunidade (AIRE-1), o gene fator necrose tumoral (TNF), entre outros, têm sido envolvidos, mas não
confirmados (19-22).
Três regiões cromossômicas podem estar ligadas à
DG, denominadas DG-1 (cromossomo 14q31), DG-2
(cromossomo 20q11.2) e DG-3 (cromossomo Xq21)
(19) e duas à TH, denominadas TH-1 (cromossomo
13q33) e TH-2 (cromossomo 12q22) (22). O complexo MHC codifica os antígenos de histocompatibilidade humana (HLA), localizado no cromossomo 6p21
(23). O alelo HLA-DR3 mostrou forte associação positiva com DG, enquanto o HLA-DR5, negativa (23).
A frequência do HLA-DR3 na população geral varia
entre 15% e 30%, enquanto em pacientes com DG, de
40% a 55%, conferindo risco relativo para pessoas com
HLA-DR3 maior que quatro (24). As associações da
DG com os haplotipos DRB1*0304, DRB1*0301 e
DQA1*0501 têm sido relatadas (23,24), inclusive no
Brasil (25), mas a exata sequência de aminoácidos na
cadeia DRβ1 de suscetibilidade à DG permanece indefinida. Recentemente, postulou-se que a associação do
DRB1*03 com DG seria resultante da presença de arginina na posição β74 do HLA-DRB1 (DRb1-Arg74),
fundamental para a capacidade dos receptores das células T fixarem e apresentarem antígenos B74 (23). Os
estudos de associação da TH com antígenos HLA são
menos consistentes, tendo sido descritas associações
Arq Bras Endocrinol Metab. 2009;53/1
com HLA-DR3 e DQB1*0301 (26) em caucasianos,
HLA-DRw53 em japoneses (27) e HLA-DR9 em chineses (28). A TH com bócio foi associada com HLADR5, enquanto a TH atrófica, com HLA-DR3 (20).
Os efeitos dessas associações entre DAIT e o complexo MHC são modestos e outros estudos mostraram
resultados conflitantes. Além disso, a suscetibilidade varia entre grupos étnicos e os estudos em gêmeos mostraram influência genética muito maior do que o HLA
individualmente (6).
A molécula CTLA-4 é o principal regulador negativo da ativação dos linfócitos T, pela competição da ligação da proteína B7 (expressa na célula apresentadora
de antígeno) à proteína coestimuladora CD28. Portanto,
mutações no gene CTLA-4 poderiam resultar ativação
exagerada dos linfócitos T e desenvolvimento de autoimunidade (29,30). O bloqueio da molécula CTLA-4 com
anticorpo monoclonal confere aumento da proliferação
das células T e da produção da interleucina-2 (30); além
disso, polimorfismos no gene CTLA-4, no cromossomo
2q33, têm sido associados com todas as formas de DAIT,
em várias populações, inclusive caucasianos e japoneses,
possivelmente em razão de ser molécula coestimuladora
inespecífica (15,22,30). Uma metanálise recente envolvendo mais de 13 mil indivíduos encontrou associação
significativa entre polimorfismo nos alelos A49G e CT60
com DG e TH (31). No entanto, embora este gene exerça papel significativo para a autoimunidade em geral, não
seria determinante no desenvolvimento da autoimunidade órgão-específica (30) e suficiente para a expressão
fenotípica de DG e TH (30,31). Assim, o papel do gene
CTLA-4 na suscetibilidade da autoimunidade tiroidiana
ainda permanece incompreendido em sua totalidade.
7
Copyright© ABE&M todos os direitos reservados.
MHC: complexo maior de histocompatibilidade; CTLA-4: antígeno-4 associado ao linfócito T citotóxico; CD40: molécula CD40 associada a células b; PTPN22: proteína tirosina
fosfatase-22; TG: tiroglobulina; TSHR: receptor do TSH; DG: doença de Graves; TH: tiroidite de Hashimoto.
Copyright© ABE&M todos os direitos reservados.
Patogênese das doenças tiroidianas autoimunes
A molécula CD40, membro da família de moléculas
do receptor do TNF (TNF-R), expressada primariamente nos linfócitos B e outras células apresentadoras de antígenos (APC), tem papel fundamental na ativação e na
proliferação dos linfócitos B (15). O gene CD40 tem sido
associado com a DG, mas não com a TH. Uma explicação
seria que um alelo C induziria a hiperexpressão da molécula CD40, resultando ativação acentuada dos linfócitos B e
predomínio da resposta imune tipo Th2 (15,19,22).
A proteína PTPN22, de modo semelhante ao CTLA-4,
é inibidor potente da ativação dos linfócitos T e polimorfismos deste gene (substituição do triptofano por arginina
no códon R620W), causam hiperativação dos linfócitos
T e também têm sido associados como determinantes
do desenvolvimento de DAIT e de múltiplos fenótipos autoimunes (22,32). No entanto, a associação de
polimorfismos neste gene com suscetibilidade à DAIT
mostra-se diferente entre as várias etnias.
Os genes específicos da glândula tiróide também
foram testados. Um candidato natural seria o gene
TSHR, mas alguns estudos têm rejeitado contundentemente esta hipótese, permanecendo indefinido se o
gene TSHR é candidato real para suscetibilidade à DG
(33). Em contraste, existem evidências sólidas do envolvimento de polimorfismo no gene TG como fator
de risco para a DAIT (34). Estudos de triagem genômica observaram forte associação entre o cromossomo
8q24 (região que contém o gene TG) e a DAIT (35).
Postula-se que variantes do gene TG poderiam iniciar
resposta autoimune tiroidiana por alterar a apresentação
do peptídeo TG pela APC às células T (22).
Em resumo, existem evidências sólidas que sugerem
forte contribuição genética na suscetibilidade para o
desenvolvimento da DAIT, mas constituída pelo efeito combinado de múltiplos genes de efeito modesto,
pois não parece haver nenhum marcador individual para
os fenótipos DG e TH. Neste modelo, genes de suscetibilidade compartilhados poderiam explicar a origem
comum de DG e TH, enquanto genes específicos explicariam as diferenças fenotípicas das doenças. Acredita-se
que a identificação de genes de grande predisposição
à DAIT e o entendimento de suas consequências funcionais resultaria melhor compreensão dos mecanismos
moleculares determinantes da DAIT, com possíveis implicações no seu tratamento e prevenção.
Determinantes ambientais de suscetibilidade
A participação de fatores ambientais na patogênese da
DAIT também tem sustentação por estudos em gêmeos
8
homozigóticos, cuja taxa de autoimunidade tiroidiana é
menor que 100% e também por modelos animais de autoimunidade tiroidiana. Além disso, estudos realizados
em populações geneticamente similares, vivendo em
condições diferentes, fortalecem esta hipótese. A taxa de
diabetes tipo 1 em crianças paquistanesas que migraram
para o Reino Unido é a mesma das residentes neste país
que não migraram e dez vezes superior à taxa encontrada no Paquistão (36). A frequência de lúpus eritematoso sistêmico é significativamente menor em africanos do
que em negros americanos, duas populações derivadas
de um mesmo grupo étnico, mas expostas a ambientes
distintos (37). Esses dados suportam consistentemente
a hipótese da exposição ambiental como fator desencadeador da autoimunidade, mas a identificação e o papel
de cada um desses fatores permanecem indefinidos.
A participação de fatores ambientais nos mecanismos de desenvolvimento da DAIT parece ocorrer desde
a vida intrauterina. A prevalência de aTPO foi maior
em estudo que envolveu mulheres com baixo peso ao
nascimento (38) e entre gêmeos homozigóticos que
nasceram com menor peso (39), sugerindo que fatores
intrauterinos associados ao baixo peso fetal seriam os
primeiros fatores de risco ambientais de suscetibilidade
para DAIT. Uma explicação seria a associação da má
nutrição fetal com menor peso esplênico e tímico, o que
poderia resultar a maturação precoce do timo e o declínio das células T supressoras.
Uma das principais características da DAIT, assim
como de outras doenças autoimunes, é sua forte preponderância no sexo feminino. A DG e a TH são, respectivamente, de cinco a dez vezes mais frequentes no
sexo feminino em relação ao masculino (2,3). A tiroidite autoimune assintomática pode ser observada em
cerca de 25% de autópsias, sendo quatro vezes mais prevalente no sexo feminino. Em estudos de rastreamento
populacional, 8% a 26% das mulheres e apenas 3% a 6%
dos homens apresentaram autoanticorpos tiroidianos
(2,3). O cromossomo X poderia estar envolvido nesta
diferença, mas o fato de a TH ser bastante prevalente
em meninas com síndrome de Turner (cariótipo X) e
pouco prevalente em meninos com síndrome de Klinefelter (cariótipo XXY) tornam esta possibilidade limitada (40), favorecendo possível efeito dos hormônios
sexuais no sistema imune, onde os estrógenos teriam
papel exacerbador e a testosterona efeito protetor. O
microquimerismo fetal (41), conceito que envolve a
transferência de células fetais para a circulação materna,
constitui um novo mecanismo que poderia estar envolvido na maior preponderância feminina. Estas células
Arq Bras Endocrinol Metab. 2009;53/1
poderiam permanecer por longo período e participar do
desencadeamento da autoimunidade tiroidiana na vida
adulta. Se isso fosse verdade, então a paridade poderia
constituir um fator de risco para DAIT, mas os resultados de recentes estudos populacionais, realizados na
Austrália (42) e na Dinamarca (43), não encontraram
associação entre paridade e DAIT. O uso de contraceptivos orais também foi envolvido como outro fator que
contribuiria para a maior prevalência de DAIT no sexo
feminino (1).
A idade parece exercer papel na patogênese da DAIT,
desde que a prevalência de autoanticorpos tiroidianos
aumente com o avançar da idade. Hawkins e cols. (44)
encontraram incidência de autoanticorpos tiroidianos de
9,8% em mulheres na Austrália, que aumentou para 15%
quando a idade superou 60 anos. O estudo de Whickham
(3), no Reino Unido, revelou prevalência de anticorpos
antimicrossomais tiroidianos de 6,8% a 9,7% em mulheres jovens e de 13,7% em mulheres com idade entre 45
a 54 anos. Acredita-se que a idade aumentaria o tempo
de exposição aos agentes meio ambientais e produziria
alterações na imunorregulação, que poderiam contribuir
no desencadeamento da tiroidite autoimune.
O estresse tem sido frequentemente associado como
fator desencadeador de doença autoimune. A relação da
DG com fatores emocionais desencadeantes foi notada desde as primeiras descrições, por Graves e Basedow
(1). A resposta hormonal ao estresse, por meio da ativação do eixo hipotálamo-hipófise-adrenal, exerce resposta imune tipo Th2, o que suprime a imunidade celular e
aumenta a humoral, explicando porque certas doenças
autoimunes são frequentemente precedidas por intenso
estresse, entre elas, a DAIT (45).
O tabagismo é fator de risco importante para a oftalmopatia da DG, mas não há forte evidência para sua
associação com DG e TH. O aumento da síntese de glicoaminoglicanos por fibroblastos do tecido retrobulbar
e o aumento da expressão de HLA-DR induzida pela
nicotina em cultura de fibroblastos orbitais são mecanismos frequentemente citados para explicar associação
com a oftalmopatia de Graves (1).
Algumas drogas têm sido implicadas no desenvolvimento de DAIT. O exemplo mais marcante foi o desenvolvimento de DG em um terço dos pacientes com
esclerose múltipla tratados com o anticorpo monoclonal
humanizado anti-CD52 (Compath – 1H), que suprime
a resposta Th1, mas não a Th2, exacerbando a resposta
humoral e a produção de anticorpos contra o TSHR
(46). Outras drogas, como os agentes retrovirais, o
IFN-α no tratamento da hepatite C, a IL-2 utilizada no
Arq Bras Endocrinol Metab. 2009;53/1
tratamento da infecção pelo vírus HIV, do carcinoma
renal metastático e do melanoma, e o fator estimulador de colônias de granulócitos e macrófagos têm sido
reportados como indutores de autoimunidade contra a
glândula tiróide (1).
Agentes infecciosos têm potencial para desencadear processo autoimune por diferentes mecanismos, entre eles o
mimetismo molecular, quando uma resposta imune a um
autoantígeno é desencadeada pela sua similaridade molecular com o antígeno estranho por meio de reação cruzada, pela ativação policlonal de linfócitos autorreativos
e pela liberação de antígenos previamente sequestrados
(1,47). No entanto, não há evidência clara da associação
de DAIT com agentes infecciosos, exceto pela presença
de aTPO em crianças portadoras de rubéola congênita
e da notável associação com a bactéria enteropatógena
Yersínia enterocolítica (YE-RP) (47-49). A prevalência
de anticorpos YE-RP foi significativamente maior em
pacientes com DG comparados ao grupo-controle e em
pacientes com TH comparados a pacientes com outras
doenças não-autoimunes da tiróide (48). Estudos experimentais sugeriram homologia antigênica entre proteínas
liberadas da YE-RP enteropatogênica e antígenos das células epiteliais tiroidianas. Esta proteína ainda permanece
não identificada, mas parece haver homologia conformacional entre a proteína hsp70 e o TSH-R, constituindo
sítio de ligação específica para o TSH e também reconhecido pelo TRAb (49).
Esses dados, embora possam sugerir o papel da infecção como desencadeador ou acelerador de DAIT,
contrastam com a crescente aceitação do que se convencionou denominar hipótese da higiene, pela qual,
o sistema imune seria educado por meio de múltiplas
e diferentes infecções, o que resultaria melhor controle da resposta imune (50). Assim, o desenvolvimento
urbano e a sensível melhora das condições de higiene,
diminuindo a exposição a agentes microbianos, poderiam estar associados com aumentado risco de doença
autoimune. Recentemente, um estudo epidemiológico
envolvendo duas populações de crianças geneticamente similares, mas expostas a diferentes meio ambientes,
uma na Finlândia e outra na Rússia, mostrou que a
prevalência de autoanticorpos tiroidianos foi significativamente maior na Finlândia, onde o nível socioeconômico e a qualidade de vida eram marcadamente
superiores (51).
O selênio (Se) é um micronutriente essencial para a
síntese de selenoproteínas que exercem papel importante na síntese, metabolismo e ação dos hormônios tiroidianos; além disso, modifica a expressão de, pelo me9
Copyright© ABE&M todos os direitos reservados.
Patogênese das doenças tiroidianas autoimunes
Copyright© ABE&M todos os direitos reservados.
Patogênese das doenças tiroidianas autoimunes
nos, 30 selenoproteínas, entre as quais, as famílias das
selenoenzimas glutationa-peroxidases, tioredoxina-peroxidases e desiodases tiroidianas (52). Possui propriedade antioxidante potente e forma sistema complexo de
defesa que protege o tirócito da lesão oxidativa (53).
A deficiência de selênio foi associada com bócio e com
hipoecogenicidade da tiróide, aspectos característicos
da TH, enquanto a suplementação com selênio parece
modificar a resposta imune, reduzindo significativamente os títulos de aTPO e a ecogenicidade tiroidiana em
pacientes com tiroidite autoimune (54). Recentemente,
observou-se que a suplementação de Se durante a gestação reduziu a incidência de disfunção tiroidiana e de
hipotiroidismo pós-parto entre as gestantes com títulos
positivos de aTPO (55).
Entre todos os candidatos a fatores ambientais de
suscetibilidade à DAIT, a concentração de iodo na dieta
assume o papel de fator exógeno principal como modulador do processo de autoimunidade tiroidiana (56).
Uma ingestão de iodo adequada é essencial para a síntese de hormônios tiroidianos e a consequente função
tiroidiana normal; entretanto, em geral, sua deficiência
atenua, enquanto o excesso de iodo acelera a indução de
tiroidite autoimune em indivíduos geneticamente suscetíveis (56-62). O papel do iodo da dieta, como fator
desencadeador de autoimunidade tiroidiana, está bem
documentado em modelos animais (57). Além disso,
estudos populacionais de monitoramento da deficiência do iodo após intervenções de sua reposição (58,59)
e a maior prevalência de DAIT em regiões suficientes
em iodo (60,61) sugerem forte associação entre a ingestão de iodo e o desenvolvimento de autoimunidade
tiroidiana. De fato, a incidência de tiroidite autoimune
nos Estados Unidos aumentou concomitante ao progressivo aumento de iodo na dieta. Na Grécia, estudo
prospectivo realizado em pacientes com bócio endêmico evidenciou aumento na incidência de autoanticorpos
tiroidianos após administração de iodo oral ou injetável
(61). A prevalência de positividade para anticorpos tiroidianos antimicrossomais foi de 25% em idosas residentes em Worcester, Massachusetts, área suficiente em
iodo, enquanto em Reggio Emilia, Itália, área deficiente
em iodo, a positividade foi menor que 1% (61). Em regiões onde a ingestão de iodo é elevada, como no Japão,
a incidência de TH é maior quando comparada às regiões onde a ingestão de iodo é normal ou relativamente
baixa (61,62). Finalmente, a DG é mais frequente em
áreas suficientes de iodo, enquanto causas não-autoimunes de hipertiroidismo são mais prevalentes em regiões
com baixo conteúdo de iodo na dieta (61).
10
Os mecanismos pelos quais o iodo da dieta modularia a reação tiroidiana autoimune são ainda desconhecidos, mas várias hipóteses têm sido aventadas, como
a toxicidade direta ao tirócito, a imunogenicidade aumentada da TG e os efeitos diretos do iodo nas células
do sistema imune (63). Assim, quantidades elevadas de
iodo são oxidadas rapidamente pela enzima TPO, gerando elementos oxidativos que podem causar lesão da
membrana celular e indução de processo inflamatório
ou autoimune em indivíduos predispostos. Por outro
lado, o tratamento com antioxidantes, de animais geneticamente predispostos para autoimunidade tiroidiana,
reduziu o infiltrado linfocítico e a síntese de anticorpos
(64). Uma ingestão excessiva de iodo resulta moléculas
de TG altamente iodadas e na formação de neoepitopos
mais imunogênicos, o que poderia precipitar o processo
autoimune. Além disso, a iodação aumentada da TG facilitaria a apresentação de epitopos patogênicos não-iodados da TG para as células apresentadoras de antígeno
(APC) (65). Finalmente, efeitos estimuladores do iodo
nas células do sistema imune, como macrófagos, linfócitos T e B, moléculas de adesão e, particularmente, nas
células dendríticas, parecem exercer papel no desencadeamento da autorreatividade tiroidiana (63).
Patogênese da autoimunidade tiroidiana
O desenvolvimento da DAIT é determinado pela perda
da tolerância imunológica e da reatividade a autoantígenos tiroidianos, resultando infiltrado na glândula por
linfócitos T e B reativos, produção de autoanticorpos
e na expressão clínica do hipertiroidismo na DG e do
hipotiroidismo na TH. Na DG, o infiltrado tiroidiano
de células T ativa as células β para a produção do anticorpo anti-receptor do TSH (TRAb), o qual ocupa e
ativa o TSHR, estimulando a tiróide e determinando
o hipertiroidismo. Por outro lado, na TH, as células T
induzem a apoptose das células foliculares e a destruição
da arquitetura glandular e hipotiroidismo (4,5). Embora inicialmente consideradas como doenças distintas, em
uma visão mais moderna e atual, DG e TH representariam lados opostos ou desfechos diferentes de um mesmo processo fisiopatológico (66).
O desenvolvimento da tolerância imunológica a autoantígenos envolve processo complexo de mecanismos
centrais e periféricos. A tolerância central ocorre no
timo pela deleção de células T que se ligam com alta
afinidade a peptídeos endógenos. Quando este processo falha, células T efetoras autorreativas (Teffs) podem
escapar da seleção tímica e migrar para a periferia, onde
Arq Bras Endocrinol Metab. 2009;53/1
são inibidas pelas células T (CD4+) naturalmente regulatórias (Treg) (67). As células Treg, geradas no timo,
expressam as moléculas CD25 e CTLA-4, consideradas
essenciais para a supressão da resposta imune mediada
por células T. Os polimorfismos do gene CTLA-4 ou
a mutação do gene CD25 associam-se com doenças
autoimunes em humanos (68) e a depleção das células
Treg tem sido relacionada com o desenvolvimento de
tiroidite autoimune, a apoptose celular e a progressão
do hipertiroidismo da DG ao hipotiroidismo da TH,
que ocorre naturalmente em alguns casos (69).
Os fatores desencadeadores do processo autoimune na DAIT não são bem conhecidos, mas admite-se
que o sinal inflamatório inicial seria emitido por lesão
ou necrose celular desencadeada por múltiplos fatores,
como anormalidades genéticas, infecção (virais ou bacterianas), estresse ou excesso de iodo, com liberação
de autoantígenos, atração e infiltração glandular por
células T e β (1,4). A lesão inicial atrairia quantidade
expressiva de APC “profissionais” para o meio intratiroidiano, que, por sua vez, apresentaria os autoantígenos tiroidianos aos linfócitos T auxiliadores CD4+. As
citocinas liberadas deste processo induziriam a expressão de moléculas MHC (HLA de classe I e classe II)
na superfície da célula folicular, transformando-as em
APCs. A expressão aberrante de moléculas HLA de classe II na célula tiroidiana parece ter papel relevante no
desenvolvimento da DAIT (23). Os mecanismos pelos
quais moléculas HLA conferem suscetibilidade à DAIT
têm sido agora mais bem compreendidos. As células T
reconhecem e respondem a um antígeno pela interação
com complexo composto de peptídeo antigênico apresentado por moléculas HLA. Especula-se que diferentes
alelos HLA tenham afinidades distintas por peptídeos
de autoantígenos tiroidianos; uma vez ligados, os peptídeos seriam apresentados e reconhecidos por receptores das células T (TCR) em células que teriam escapado da tolerância imunológica. Neste modelo, um alelo
HLA-DR específico poderia permitir que um peptídeo
autoantigênico se fixe, seja apresentado e reconhecido
pelo TCR (23). Neste sentido, tem sido demonstrado
que a presença da arginina na posição 74 da cadeia DR
β1 do HLA-DR3 (DRb1-Arg74) induziria alteração estrutural da unidade de ligação de peptídeos do HLADR, afetando, de modo significativo, sua capacidade de
ligação a peptídeos tiroidianos patogênicos (23,66,68).
Em uma outra alternativa, agentes infecciosos desencadeariam o processo autoimune por mecanismo conhecido por mimetismo molecular, quando resposta imune
a autoantígeno fosse desencadeada pela sua similaridade
Arq Bras Endocrinol Metab. 2009;53/1
molecular com o antígeno estranho por meio de reação
cruzada (47-49).
Havendo falha na manutenção da tolerância imunológica, os autoantígenos não seriam reconhecidos, resultando ativação de células β e T autorreativas, com
resposta inflamatória excessiva e inapropriada. O recrutamento de linfócitos na DAIT envolve processo complexo com atuação de moléculas de adesão e, principalmente, de quimiocinas, uma família especializada de
citoninas que controlam a migração de leucócitos (quimiotaxia) durante o processo inflamatório (70). Estudos
experimentais sugerem que as quimiocinas induzidas
pelo IFN-γ (CXCL9, CXCL10, CXCL11) e seu receptor CXCR3 teriam papel importante no estágio inicial
da DAIT, uma vez que essas quimiocinas recrutariam
linfócitos Th1, que secretam IFN-γ, portanto perpetuando o processo autoimune. Por outro lado, os linfócitos Th2 são recrutados nos tecidos pelas quimiocinas
CCL17 e CCL22, ligantes do receptor CCR4, expressos nas células Th2 (70). As células Th1 secretam IL-2,
IFN-γ e TNF-a, que resulta na ativação de macrófagos,
na produção de fixadores do complemento, em anticorpos opsonizantes e em citotoxicidade. As células Th2
secretam IL-4, IL-5, IL-6, IL-10 e IL-13, que têm papel inibitório sobre a produção das citocinas Th1 e estimulam os linfócitos B na produção de imunoglobulinas
específicas (71-72). Assim, as quimiocinas poderiam ter
papel importante no tipo de resposta linfocítica predominante, se Th1 ou Th2 (70). A diferenciação em uma
ou outra resposta imune parece, ainda, ser regulada por
sinais coestimuladores determinados pela família de moléculas expressas na superfície celular das células APC,
denominadas proteínas B7. Coestimuladores B7-1 induzem a produção de células Th1, enquanto proteínas
coestimuladoras B7-2 induzem a produção de células
Th2 (71-73).
Na TH, a maioria do infiltrado linfocítico age como
células Th1, favorecendo a imunidade celular e o desenvolvimento da apoptose celular (74). Ligantes apoptóticos e receptores, como o TNF, Fas e o ligante indutor de apoptose ligado a necrore tumoral (TRAIL)
são expressados na célula tiroidiana, mas, em condições
fisiológicas, permanecem inativados (74). No entanto, a
expressão do FasL, induzidas por citocinas Th1 no infiltrato linfocítico tiroidiano, determina a apoptose (75).
Defeitos nas células Treg resultam a hiperprodução das
citocinas Th1 e poderiam estar envolvidos na patogênese da TH (76).
Na DG, o predomínio de citocinas Th2 favorece a
imunidade humoral com a produção aumentada de au11
Copyright© ABE&M todos os direitos reservados.
Patogênese das doenças tiroidianas autoimunes
Patogênese das doenças tiroidianas autoimunes
toanticorpos pelos linfócitos B. O aumento da concentração da imunoglobulina G (IgG) ou as citocinas Th2
parecem inibir a expressão de Fas e induzir a expressão de
moléculas antiapoptóticas, o que protegeria os tirócitos
contra a apoptose na DG (76). Entretanto, em modelos
animais, a produção do TRAb foi associada tanto com
a resposta tipo Th1 quanto Th2, sugerindo que a DG
possa envolver diferentes tipos de resposta imune (77).
Um novo subtipo de resposta Th, denominado Th17,
também poderia estar envolvido na patogênese da DG
(77). As células Th17 desenvolvem-se em resposta às
citocinas IL-23, IL-6 e TGFβ1 por células dendríticas e
antagonizam as respostas tipo Th1 e Th2 (70).
Concluindo, a DAIT é o resultado da interação entre múltiplos fatores ambientais e múltiplos genes, com
importância variável na indução da autoimunidade em
um indivíduo ou em uma população. Em outras palavras, fator ambiental específico, na presença de gene de
suscetibilidade, poderia ser determinante na indução da
autoimunidade tiroidiana em um indivíduo, enquanto
a interação entre um segundo fator ambiental e um diferente gene de suscetibilidade, precipitaria o início da
autoimunidade em outro. Acredita-se que conhecimento
mais preciso dos mecanismos de interação entre fatores
ambientais e genes na indução da autoimunidade tiroidiana, possivelmente resultará o desenvolvimento de estratégias de prevenção em uma determinada população.
A Figura 1 sumariza os principais passos dos mecanismos de suscetibilidade, desenvolvimento e progressão da autoimunidade tiroidiana.
Agradecimentos: A parte experimental do trabalho dos autores na área
de autoimunidade tiroidiana é financiada pela Fundação de Amparo à
Pesquisa do Estado de São Paulo (Fapesp) (Auxílio nº 06/59737-9).
RMBM é pesquisador do Conselho Nacional de Desenvolvimento
Científico e Tecnológico (CNPq), do Fundo de Auxílio aos Docentes
e Alunos (FADA)/Unifesp e do Instituto Fleury. Os autores agradecem a excelente assistência secretarial de Ângela Faria.
Declaração: Os autores declaram não haver conflitos de interesse
científico neste artigo.
Suscetibilidade
genética
Fator ambiental
(Iodo, toxina viral, bacteriana etc.)
Lesão celular
Liberação de autoantígenos
Apresentação de autoantígenos pelas células APC
Perda da autotolerância imunológica;
resposta imune inapropriada e exacerbada
Copyright© ABE&M todos os direitos reservados.
Infiltrado linfocítico intratiroidiano
Resposta predominante Th1
Imunidade celular e apoptose
Tiroidite de Hashimoto
Resposta predominante Th2
Imunidade humoral, produção de TRAb
Doença de Graves
Figura 1. Patogênese das doenças tiroidianas autoimunes.
12
Arq Bras Endocrinol Metab. 2009;53/1
Patogênese das doenças tiroidianas autoimunes
1. Prummel MF, Strieder T, Wiersinga WM. The environment and autoimmune thyroid diseases. Eur J Endocrinol. 2004;150:605-18.
2. Wang C, Crapo L. The epidemiology of thyroid disease and implications
for screening. Endocrinol Metab Clin North Am. 1997;26:189-219.
3. Tunbridge WM, Vanderpump MPJ. Population screening for autoimmune thyroid disease. Endcrinol Metab Clin North Am. 2000;
29:239-53.
4. Weetman AP. Autoimmune thyroid disease: propagation and progression. Eur J Endocrinol. 2003;148:1-9.
5. Collins J, Gough S. Autoimmunity in thyroid disease. Eur J Nucl Med.
2002;29 Suppl 2: S417-24.
6. Lazarus JH, Parkes AB, Premawardhana LD. Postpartum thyroiditis.
Autoimmunity. 2002;35:169-73.
7. Woolf PD. Transient painless thyroiditis with hyperthyroidism: a variant of lymphocytic thyroiditis? Endocr Rev. 1980;1:411-20.
8. Marazuela M, Garcia-Buey L, Gonzalez-Fernandez B, Garcia-Monzon C,
Arranz A, Borque MJ, et al. Thyroid autoimmune disorders in patients
with chronic hepatitis C before and during interferon-a therapy. Clin
Endocrinol (Oxf). 1996;44:635-42.
23.Jacobson EM. Huber A, Tomer Y. The HLA gene complex in thyroid
autoimmunity: from epidemiology to etiology. J Autoimmun. 2008;
30:58-62.
24.Yanagawa T, Mangklabruks A, DeGroot LJ. Strong association between
HLA-DQA1*0501 and Grave’s disease in a male Caucasian population.
J Clin Endocrinol Metab. 1994;79:227-9.
25.Maciel LM, Rodrigues SS, Dibbern RS, Navarro PA, Donadi EA. Association of the HLA-DRB1*0301 and HLA-DQA1*0501 alleles with Graves’
disease in a population representing the gene contribution from several ethnic backgrounds. Thyroid. 2001;11:31-5.
26.Wu Z, Stephen HAF, Sachs JA, Biro PA, Cutbush S, Magzoub MM, et
al. Molecular analysis of HLA-DQ and DP genes in Caucasoid patients
with Hashimoto’s thyroiditis. Tissue Antigens. 1994;43:116-9.
27.Honda K, Tamai H, Morita T, Kuma K, Nishimura Y, Sasazuki T.
Hashimoto’s thyroiditis and HLA in japanese. J Clin Endocrinol Metab.
1989;69:1268-73.
28.Hawkins BR, Lam KSL, Ma JTC, Wang C, Yeung RTT. Strong association between HLA-DRw9 and Hashimoto’s thyroiditis in Southern Chinese. Acta Endocrinol (Copenh).1987;114:543-6.
29.Chistiakov DA, Turakulov RI. CTLA-4 and its role in autoimmune thyroid disease. J Mol Endocrinol. 2003;31:21-36.
9. Dittmar M, Kahaly GJ. Polyglandular autoimmune syndromes: immunogenetics and long-term follow-up. J Clin Endocrinol Metab.
2003;52:759-64.
30.Tomer Y, Greenberg DA, Barbesino G, Concepcion E, Davies TF. CTLA4 and not CD28 is a susceptibility gene for thyroid autoantibody production. J Clin Endocrinol Metab. 2001;86:1687-93.
10.Raporport B, McLachlan SM. Thyroid autoimmunity. J Clin Invest.
2001;108:1253-9.
31.Kavvoura FK, Akamimizu T, Awata T, Ban Y, Chistiakov DA, Frydecka
I, et al. Cytotoxic T-lymphocyte associated antigen 4 gene polymorohisms and autoimmune thyroid disease: A meta-analysis. J Clin Endocrinol Metab. 2007;92:3162-70.
11.McLachlan S, Rapoport B. Thyroid peroxidase as an autoantigen. Thyroid. 2007;17:939-48.
12.Smith BR, Sanders J, Furmaniak J. TSH receptor antibodies. Thyroid.
2007;17:923-37.
13.Brix TH, Kyvik KO, Chridtensen K, Hegedus L. A population-based
study of chronic autoimmune hypothyroidism in Danish twins. J Clin
Endocrinol Metab. 2000;85:536-9.
14.Brix TH, Kyvik KO, Chridtensen K, Hegedus L. Evidence for a major role
of heredity in Graves’ disease: a population-based study of two Danish
twin cohorts. J Clin Endocrinol Metab. 2001;86:930-4.
15.Jacobson EM, Tomer Y. The CD40, CTLA-4, thyroglobulin, TSH receptor, and PTPN22 gene quintet and its contribution to thyroid autoimmunity: back to the future. J Autoimmun. 2007;28:85-98.
16.Hansen PS, Brix TH, Iachine I, Kyvik KO, Hegedus L. The relative importance of genetic and environmental effects for the early stages of
thyroid autoimmunity: a study of healthy Danish twins. Eur J Endocrinol. 2006;154:29-38.
17.Brix TH, Kyvik KO, Hagedus L. What is the evidence of genetic factors in the etiology of Graves’ disease? A brief review. Thyroid. 1998;
8:727-34.
32.Criswell LA, Pfeifer KA, Lum RF, Gonzales B, Novitzke J, Kern M, et al.
Analysis of families in the multiple autoimmune disease genetics consortium (MADGC) collection: the PTPB22 620W allele associates with
multiple autoimmune phenotypes. Am J Hum Genet. 2005;76:561-71.
33.Roux N, Shields DC, Misrahi M, Ratanachaiyavong S, McGregor AM,
Milgrom E. Analysis of the TSH receptor as a candidate gene in familial
Graves’ disease. J Clin Endocrinol Metab. 1996;81:3483-6.
34.Tomer Y, Greenberg DA, Concepcion E, Ban Y, Davies TF. Thyroglobulin is a thyroid specific gene for the familial autoimmune thyroid diseases. J Clin Endocrinol Metab. 2002;87:404-7.
35.Taylor JC, Gough SC, Hunt PJ, Brix TH, Chatterjee JM, Connell JA,
et al. A genome-wide screen in 1119 relative pairs with autoimmune
thyroid disease. J Clin Endocrinol Metab. 2006;91:646-53.
36.Bodansky HJ, Staines A, Stephenson C, Haigh D, Cartwright R. Evidence for an environmental effect in the etiology of insulin dependent
diabetes in a transmigratory populations. BMJ. 1992;304;1020-2.
37.Symmons DPM. Frequency of lupus in people of African origin. Lupus.
1995;4:176-8.
18.Hall R, Stanbury JB. Familial studies of autoimmune thyroiditis. Clin
Exp Immunol. 1967;2:719-25.
38.Philips DI, Cooper C, Fall C, Prentice L, Osmond C, Barker DJ, et al. Fetal
growth and autoimmune thyroid disease. Q J Med. 1993;86:247-53.
19.Jacobson EM, Tomer Y. The genetic basis of thyroid autoimmunity.
Thyroid. 2007;17:949-61.
39.Phillips DI, Osmond C, Baird J, Huckle A, Rees-Smith B. Is birth weight
associated with thyroid autoimmunity? A study in twins. Thyroid.
2002;12:377-80.
20.Tomer Y, Ban Y, Concepcion E, Barbesino G, Villanueva R, Greenberg
DA, et al. Commom and unique susceptibility loci in Graves and Hashimoto diseases: results of whole-genome screening in a data set of 102
multiplex families. Am J Hum Genet. 2003;73:736-47.
21.Ban Y, Davies TF, Greenberg DA, Concepcion ES, Tomer Y. The influence of human leucocyte antigen (HLA) genes on autoimmune thyroid
disease (AITD): results of studies in HLA-DR3 posotive IITD families.
Clin Endocrinol (Oxf). 2002;57:81-8.
22.Tomer Y, Davies TF. Searching for the autoimmune thyroid disease
susceptibility genes: from gene mapping to gene function. Endocr Rev.
2003;24:694-717.
Arq Bras Endocrinol Metab. 2009;53/1
40.Barbesino G, Chiovato L. The genetics of Hashimoto’s disease. Endocrinol Metab Clin North Am. 2000;29:357-74.
41.Klintschar M, Schwaiger P, Mannweiler S, Regauer S, Kleiber M. Evidence of fetal microchimerism in Hashimoto’s thyroiditis. J Clin Endocrinol
Metab. 2001;86:2494-8.
42.Walsh JP, Bremner AP, Bulsara MK, O’Leary P, Leedman PJ, Feddema P,
et al. Parity and the risk of autoimmune thyroid disease: a communitybased study. J Clin Endocrinol Metab. 2005;90:5309-12.
43.Pedersen IB, Lauberg P, Knudsen N, Jorgensen T, Perrild H, et al. Lack of
association between thyroid autoantibodies and parity in a population
13
Copyright© ABE&M todos os direitos reservados.
REFERÊNCIAS
Patogênese das doenças tiroidianas autoimunes
study argues against microchimerism as a trigger of thyroid autoimmunity. Eur J Endocrinol. 2006;154:39-45.
59.Teng W, Shan Z, Teng X, Guan H, Li Y, Jin Y, et al. Effect of iodine intake
on thyroid diseases in China. N Engl J Med. 2006;354:2783-93.
44.Hawkins BR, Dawkins RI, Burger HG, et al. Diagnostic significance of
thyroid microssomal antibodies in a randomly selected population.
Lancet. 1980;2:1057-61.
60.Konno N, Makita H, Yuri K, Iizuka N, Kawasaki K. Association between
dietary iodine intake and prevalence of subclinical hypothyroidism in
the coastal regions of Japan. J Clin Endocrinol Metab. 1994;78:393-7.
45.Elenkov IJ, Chrousos GP. Stress hormones, proinflamatory and anti-inflammatory cytokines, and autoimmunity. Ann N Y Acad Sci.
2002;966:290-303.
61.Papanastasiou L, Valalas IA, Koutras DA, Mastorakos G. Thyroid autoimmunity in the current iodine environment. Thyroid. 2007;17:729-39.
46.Coles AJ, Wing M, Amith S, Coraddu F, Greer S, Taylor C, et al. Pulsed
monoclonal antibody treatment and autoimmune thyroid disease in
multiple sclerosis. Lancet. 1999;354:1694-5.
47.Tomer Y, Davies TF. Infection, thyroid disease, and autoimmunity. Endocr Rev. 1993;14:107-20.
48.Weiss M, Ingbar SH, Winblad S, Kasper DL. Demonstration of a saturable binding site for thyrotropin in yersinia enterocolitica. Science.
1983;219:1331-3.
49.Heyma P, Harrison LC, Robins-Brownw R. Thyrotrophin (TSH) binding
sites on Yersinia enterocolitica recognized by immunoglobulins from
humans with Graves’ disease. Clin Exp Immunol. 1986;64:249-55.
50.Bloomfield SF, Stanwell-Smith R, Crevel RW, Pickup J. Too clean, or
not too clean: the hygiene hypothesis and home hygiene. Clin Exp Immunol. 2006;36:402-25.
51.Kondrashova A, Viskari H, Haapala Am, Seiskari T, Kulmala P, LLonen
J, et al. Serological evidence of thyroid autoimmunity among schoolchildren in two different socioeconomic environments. J Clin Endocrinol Metabol. 2008;93;729-34.
52.Dumont JE. Selenium, the thyroid, and the endocrine system. Endocr
Rev. 2005;26:944-84.
53.Duntas LH. The role of selenium in thyroid autoimmunity and cancer.
Thyroid. 2006;16:455-60.
54.Gaertner R, Gasnier BC, dietrich JW, Krebs B, Angstwurm MW. Selenium supplementation in patients with autoimmune thyroidites
decreases thyroid peroxidase antibodies concentrations. J Clin Endocrinol Metab. 2002;87:1687-91.
55.Negro R, Greco G, Mangieri T, Pezzarossa A, Dazzi D, Hassan H. The
influence of selenium supplementation on pospartum thyroid status in
pregnant women with thyroid peroxidase autoantibodies. J Clin Endocrinol Metab. 2007;92:1263-8.
56.Rose NR, Rasooly L, Saboori AM, Burek CL. Linking iodine with autoimmune thyroiditis. Environ Health Perspect. 1999;107 Suppl 5:749-52.
57.Ruwhof C, Drexhage HA. Iodine and thyroid autoimmune disease in
animal models. Thyroid. 2001;11:427-36.
63.Rose NR, Bonita R, Burek CL. Iodine: an environmental trigger of
thyroiditis. Autoimmun Rev. 2002;1:97-103.
64.Bagchi N, Brown TR, Sundeck RS. Thyroid cell injury is an initial event
in the induction of autoimmune thyroiditis by iodine in obese strain
chickens. Endocrinology. 1995;136:5054-60.
65.Carayanniotis G. Recognition of thyroglobulin by T cells: the role of
iodine. Thyroid. 2007;17:963-73.
66.McLachlan SM, Nagayama Y, Pichurin PN, Mizutori Y, Chen CR, Misharin
A, et al. The link between Graves’ disease and Hashimoto’s thyroiditis: a
role for regulatory T cells. Endocrinology. 2007;148:5724-93.
67.Sakaguchi S, Yamaguchi T, Nomura T, Ono M. Regulatory T cells and
immune tolerance. Cell. 2008;133:775-87.
68.Zeitin AA, Simmonds MJ, Gough SCL. Genetic developments in autoimmune thyroid disease: an evolutionary process. Clin Endocrinol.
2008;68:671-82.
69.Marazuela M, Garcia-López MA, Figueroa-Vega N, Fuente H, AlvaradoSánchez B, Monsiváis-Urenda A, et al. Regulatory T cells in human autoimmune thyroid disease. J Clin Endocrinol Metab. 2006;91:3639-46.
70.Rotondi M, Chiovato L, Romagnani S, Serio M, Romagnani P. Role
of chemokines in endocrine autoimmune diseases. Endocr Rev.
2007;28:492-520.
71.Berger A. Th1 and Th2 responses: what are they? BMJ. 2000;321:424.
72.Romaldini JH, Sgarbi JA. Tiroidites. In: Saad MJA, Maciel RMB, Mendonça BB, editors. Endocrinologia. São Paulo, Atheneu; 2007. p. 413-22.
73.Stelios F, Agathocles T. On the pathogenesis of autoimmune thyroid
disease: a unifying hypothesis. Clin Endocrinol. 2004;60:397-409.
74.Wang SH, Baker Jr JR. The role of apoptosis in thyroid autoimmunity.
Thyroid. 2007;10:975-9.
75.Giordano C, Stassi G, De Maria R, Todaro M, Richiusa P, Papoff G, et
al. Potential involvement of Fas and its ligand in the pathogenesis of
Hashimoto’s thyroiditis. Science. 1997;275:960-3.
76.Bretz JD, Baker JR Jr. Apoptosis and autoimmune thyroid disease: following a TRAIL to thyroid destruction? Clin Endocrinol. 2001;55:1-11.
77.Nagayama Y. Graves’ animal models of Graves’ hyperthyroidism.
Thyroid. 2007;10:981-7.
Copyright© ABE&M todos os direitos reservados.
58.Zois C, Stavrou I, Svarna E, Seferiadis K, Tsatsoulis A. Natural course
of autoimmune thyroiditis after elimination of iodine deficiency in northwestern Greece. Thyroid. 2006;16:289-93.
62.Laurberg P, Pedersen B, Knudsen N, Ovesen L, Andersen S. Environmental iodine intake affects the type of nonmalignant thyroid disease.
Thyroid. 2001;11:457-69.
14
Arq Bras Endocrinol Metab. 2009;53/1
ANEXO 2
Trabalho No. 02
Parity is not related to autoimmune thyroid disease in a population-based study
of Japanese-Brazilians
José A. Sgarbi, Teresa S. Kasamatsu, Luiza K. Matsumura, Rui M.B. Maciel
“Thyroid”
(Thyroid 2010; 20: 1151 - 1156)
42
THYROID
Volume 20, Number 10, 2010
ª Mary Ann Liebert, Inc.
DOI: 10.1089/thy.2009.0424
IMMUNOLOGY, AUTOIMMUNITY, AND GRAVES’ OPHTHALMOPATHY
Parity Is Not Related to Autoimmune Thyroid Disease
in a Population-Based Study of Japanese-Brazilians
José A. Sgarbi,1,2 Teresa S. Kasamatsu,1 Luiza K. Matsumura,1 and Rui M.B. Maciel1
Background: It has been suggested that the female preponderance for autoimmune thyroid disease might be
associated with hormonal differences, abortion, and fetal microchimerism. Findings emerging from the few
epidemiological studies on this matter, however, are controversial. In this study, we investigated the hypothesis
whether parity, abortion, and the use of estrogens are associated with a higher risk for thyroid autoimmunity.
Methods: This cross-sectional population-based study examined 675 women from a Japanese-Brazilian population aged above 30 years. Thyroid peroxidase antibodies (TPOAbs), thyroglobulin antibodies (TgAbs), thyrotropin, and free T4 were measured by immunofluorimetric assays. Urinary iodine concentration was measured
using a colorimetric method. Data were analyzed in logistical regression models to obtain the odds ratio (OR)
and 95% confidence intervals.
Results: TPOAbs and TgAbs were present in 11.6% and 13.6% of women, respectively. After adjustment for age,
smoking, and urinary iodine concentration, the OR for positive TPOAb (OR, 1.22 [95% confidence interval, 0.73–
2.02]) and for positive TgAb (OR, 1.01 [0.63–1.62]) among women who had one or more parities did not differ
from those who had never given birth. In addition, we found no association between the presence of thyroid
antibodies and previous abortions or the use of estrogens.
Conclusions: Parity, abortion, and the use of estrogens are not associated with thyroid autoimmunity in this
population. These findings reinforce previous reports that advocated against a key role of fetal microchimerism
in the pathogenesis of autoimmune thyroid disease.
Introduction
A
utoimmune thyroid disease (AITD) is one of the most
common autoimmune disorders, affecting around 5% of
iodine-sufficient populations (1,2). Interaction of genetic susceptibility and environmental factors appears to be of fundamental importance in initiating the process of thyroid
autoimmunity (3). It has been postulated that 79% of the AITD
susceptibility can be attributed to genetic factors, whereas
21% can be attributed to environmental factors (4). However,
there is no clear evidence of causality, and the mechanisms by
which environmental factors trigger thyroid autoimmunity in
genetically predisposed individuals remain unclear (3,4).
Similar to other human autoimmune disorders, AITD
preferentially develops in women of childbearing age and
appears to be modulated by pregnancy (5). Possible reasons
for this and for the female preponderance for AITD might
include X-chromosomal factors, the potential effect of exogenous estrogens use on the immune system, and abortions
(3,4,6). In addition, a phenomenon known as microchimerism,
defined as a bi-directional trafficking of maternal and fetal
cells during pregnancy, has emerged to explain the female
preponderance in AITD (7–8). The persistence of microchimeric fetal cells in maternal tissues, such as thyroid, skin,
and pancreas (9,10), is believed to play a role in the pathogenesis of a number of autoimmune diseases (e.g., AITD,
systemic sclerosis, and type I diabetes mellitus) (8–10).
Whether the last factor is natural or pathogenic and whether it may be involved in the pathogenesis of AITD is still not
clear. Studies of fetal-maternal microchimerism in the thyroid
have documented a higher prevalence of fetal cells in association with Hashimoto’s thyroiditis and Graves’ disease when
compared with nonautoimmune thyroid disorders (11–13).
However, recent data emerging from epidemiological studies
show conflicting results (14–16). Two previous populationbased studies (14,15) found no association of pregnancy and
parity with thyroid antibodies or thyroid dysfunction, suggesting no role of fetal microchimerism in AITD, whereas
1
Laboratory of Molecular Endocrinology, Division of Endocrinology, Department of Medicine, Escola Paulista de Medicina, Federal
University of São Paulo, São Paulo, Brazil.
2
Division of Endocrinology, Department of Medicine, Faculdade de Medicina de Marilia, Marilia, Brazil.
1151
1152
another study demonstrated that parity was a potential risk
factor for AITD (16).
In the present study, we estimated the influence of previous
parities, abortions, and the use of estrogens on the risk of
thyroid autoimmunity in an entire community.
Materials and Methods
The present population (n ¼ 675) represents the female
portion of a nonmixed Japanese-Brazilian population living
in Bauru (Human Development Index, HDI 0.825; www
.ipeadata.com.br Accessed May 20, 2010), State of São Paulo,
Brazil. Detailed descriptions of this survey have been previously reported (17,18). Briefly, the entire population 30 years
of age (n ¼ 1751) was invited to participate, and 1330 (76%)
agreed. We excluded individuals who self-reported previous
thyroid disease or were taking thyroid drugs (n ¼ 47), those
using drugs that could affect thyroid function such as amiodarone, lithium, or corticosteroids (n ¼ 6) and individuals for
whom serum was not available for testing for thyroid antibodies (n ¼ 5). Of the remaining 1272 (95.6%) individuals, 675
women took part in the present analysis.
The participants answered a standardized questionnaire
including information concerning family and personal history
of thyroid disease, concomitant medications, smoking habits,
pregnancy, parity, and previous or present use of estrogens.
Fasting blood and urine samples were collected and stored
at 808C for analysis of thyrotropin (TSH), free T4 (FT4),
thyroid antibodies, and urinary iodine concentration (UIC).
Serum TSH was measured in duplicate by a sensitive immunofluorimetric assay (Wallac-Delphia). The functional
sensitivity of the assay was 0.05 mU/L, and the reference
range was 0.4–4.5 mU/L. Serum FT4 was measured using a
competitive immunoassay (Wallac–Delphia, Finland), and
the normal reference range was 0.7–1.5 ng/dL.
Serum antithyroid peroxidase antibodies (TPOAbs) were
determined by a time-resolved immunofluorimetric assay
(AutoDelfia TPOAb; PerkinElmer Life and Analytical Sciences, Wallac Oy). Tests were considered positive when values were above 35 U/mL, which corresponded to the
functional sensitivity given by the manufacturer. Serum antithyroglobulin antibody (TgAb) concentrations were measured using an in-house immunofluorimetric assay (19) with a
sensitivity of 40 UI/mL and an interassay error less than 5%
along the standard curve.
UIC was measured in early morning urine samples using a
modified colorimetric semiautomated method that was previously described (20). The detection limit for the method was
10 mg/L with a normal range between 100 and 299 mg/L.
Euthyroidism was defined as both serum TSH and FT4
within the normal reference range, subclinical hyperthyroidism as a TSH level less than 0.45 mU/L with normal FT4 level,
overt hyperthyroidism as a TSH level less than 0.1 mU/L in
the presence of an FT4 level above the upper limit of the
normal reference range, subclinical hypothyroidism as a TSH
level above 4.5 mU/L with a normal FT4 level, and overt
hypothyroidism as a TSH level above 4.5 mU/L with an FT4
level below the inferior limit of the normal reference range or a
TSH concentration above 20 mU/L.
This study was approved by the Ethics Committee of Escola
Paulista de Medicina, Federal University of São Paulo, and
written informed consent was obtained from all participants.
SGARBI ET AL.
Statistical analysis
All statistical analyses were performed using SAS statistical
software version 9.1 (SAS Institute Inc.). The assumed level of
significance was p < 0.05 (2-tailed). Parity was analyzed both
as a binary variable (0, 1) and as a continuous variable (0, 1,
2, and 3). Continuous variables were described with
the mean and standard deviation and nominal variables with
the absolute (n) or relative (%) frequencies according to the
number of parities, abortions, and use of estrogen. The
Kruskal-Wallis test was used for continuous data followed by
the Mann-Whitney U test when it was significant. Variables
without a normal distribution as determined by the Kolmogorov-Smirnov test underwent logarithmic transformations
before statistical analysis. Frequencies were compared by the
chi-square test or Fisher test when more than 20% of expected
values were lower than 5 or any expected value was less than
1. Logistic regression models, adjusted for age, UIC and
smoking, were used to determine the association of parities,
abortions, and the use of estrogens with the presence of thyroid antibodies; regression models were also used to determine the odds ratio and 95% confidence interval.
Results
Among the 675 women from this population, 367 (54.4%)
had never given birth and 308 (45.6%) mentioned previous
childbirths (range, 1–12). One or more abortions were disclosed by 35 (5.2%) participants, and 67 (9.9%) had previously
or were presently using estrogens (oral contraceptives or
hormone replacement therapy). The characteristics of these
women are shown in Table 1. Women who had never given
birth tended to be older ( p < 0.0001) and were more likely to
be smokers ( p < 0.04) and to have a lower UIC ( p ¼ 0.005)
than those who had at least one parity. Conversely, there were
no apparent differences in the prevalence of goiter or the
mean TSH and FT4 levels among women according to their
obstetrical history.
The median UIC for all women was 210 mg/dL, and no
statistical difference was found among thyroid disease categories. The greatest proportion of the participants had euthyroidism (80.1%), and the most prevalent thyroid
dysfunctions were subclinical hypothyroidism (10.4%) and
subclinical hyperthyroidism (6.9%). Overt hyperthyroidism
(1.6%) was more prevalent than overt hypothyroidism (1%).
We found no association of thyroid status with parity, abortion, or use of estrogens (Table 2).
Thyroid antibodies (TPOAb and/or TgAb) were positive in
17.9% of the participants (Table 1). Both tests were positive in
47 (6.9%) individuals, TPOAb was positive in 76 (11.3%) individuals, and TgAb was positive in 92 (13.6%) individuals.
There was no statistical difference in the prevalence of positive thyroid antibodies between women who had never been
given birth and those with one or more previous parity. We
repeated the analysis using number of parity as a continuous
variable (0, 1, 2, 3), but again no relationship was found
between parity and thyroid autoimmunity (data not shown).
In addition, no difference was found in the prevalence of
positive thyroid antibodies for participants who reported a
previous abortion or for those exposed to estrogens (present
or previous) through oral contraceptives or hormone replacement therapy (Table 1). However, when we analyzed
these data by quantitative levels of thyroid antibodies rather
1153
Yes (n ¼ 67)
54.0 8.2
5 (7.5)
11 (16.4)
210 196
2.9 11.2
1.09 0.6
31.9 137.3
58.8 324.6
5 (7.5)
8 (11.9)
4 (6.1)
9 (13.4)
No (n ¼ 608)
57.4 12.6
75 (12.3)
80 (13.2)
200 182
2.1 1.9
1.05 0.37
25.6 109.3
47.9 222.8
71 (11.7)
84 (13.8)
43 (7.1)
112 (18.4)
0.01
0.24
0.45
0.96
0.6
0.2
0.8
0.95
0.3
0.9
1.0
0.31
than by dichotomous classification, we found significantly
increased mean serum levels of both TPOAb and TgAb in
women who suffered one or more abortions.
As shown in Table 3, the risk (odds ratio) for positive
TPOAb and/or for positive TgAb was not different between
the following groups: (a) women who had never given birth
and those who had one or more previous childbirths; (b)
women who had a previous abortion and those who never
had an abortion; (c) women who reported present or previous
use of estrogens and those who had not previously used estrogens. After adjustment for age, UIC, and smoking, the risk
for positive thyroid antibodies (TPOAb and/or TgAb) remained unchanged among the groups.
52.9 10.3
28 (9.1)
43 (14)
220 201
2.6 4.7
1.08 0.5
24.4 90.5
56.3 274.9
38 (12.3)
41 (13.3)
21 (6.9)
58 (18.8)
60.5 12.7
52 (14.2)
48 (13.1)
190 187
2.9 13.7
1.1 0.7
27.7 127.9
42.8 194.2
38 (10.4)
51 (14.0)
26 (7.2)
63 (17.2)
Age, years
Smokers, n (%)
Goiter, n (%)
Urinary iodine concentration, mg/L
Thyrotropin, mU/L
Free T4, ng/dL
TPOAb, U/mL
TgAb, UI/mL
Positive TPOAb, n (%)
Positive TgAb, n (%)
Positive TPOAb and TgAb, n (%)
Positive TPOAb and/or TgAb, n (%)
Data presented as mean SD, unless noted otherwise.
TPOAb, thyroid peroxidase antibody; TgAb, thyroglobulin antibody.
1 (n ¼ 308)
0 (n ¼ 367)
<0.0001
0.04
0.73
0.005
0.7
0.054
0.23
0.9
0.42
0.8
0.87
0.6
57.2 12.4
74 (11.6)
88 (13.8)
210 198
2.7 10.7
1.09 0.6
15.5 52.6
28.9 91.5
72 (11.3)
87 (13.6)
43 (6.7)
116 (18.1)
53.7 10.3
6 (17.1)
3 (8.6)
185 149
3.6 7.4
1.0 0.2
26.7 114.8
49.9 240.2
3 (8.6)
4 (11.4)
3 (8.6)
4 (11.4)
0.07
0.32
0.6
0.23
0.54
0.17
0.007
0.018
0.78
1.0
0.72
0.29
Discussion
Parameter
p-Value
0 (n ¼ 640)
1 (n ¼ 35)
p-Value
Use of estrogens
No. of abortions
No. of parities
Table 1. Sociodemographic and Thyroid-Related Characteristics, According the Number of Parities, Abortions, and the Use of Estrogens
p-Value
PARITY IS NOT RELATED TO AITD
In this iodine-sufficient population of Japanese-Brazilians,
we found no relationship between previous parity and the
presence of thyroid antibodies (TPOAb and/or TgAb), thyroid dysfunction, or goiter. Our results are in close agreement
with two previous larger population-based studies (14,15), in
which previous pregnancy or parity were not risk factors for
thyroid autoimmunity. Walsh et al. (14) examined available
serum samples from 1045 of 2142 female participants in the
context of the Busselton Health Study in Western Australia,
which is considered an iodine-sufficient region. Hypo- and
hyperthyroidism (including subclinical and overt dysfunctions) were prevalent in 9% and 4% of the participants, respectively, and thyroid antibodies (TPOAb and/or TgAb)
were prevalent in 20%. After adjustment for age, no evidence
was found for increased risk of thyroid autoimmunity or abnormal TSH with increasing number of reported pregnancies.
Pedersen and colleagues (15) examined 3283 women randomly selected from the general population of Aalborg and
Copenhagen as part of the Danish investigation of iodine intake and thyroid diseases (DanThyr). With median UICs of
45 mg/L (Aalborg) and 61 mg/L (Copenhagen), the overall
prevalence for TPOAb and/or TgAb was 20.9%. Again, no
association was found between the risk of having thyroid
antibodies and a previous pregnancy, number of pregnancies,
parity, previous abortion, or the use of estrogens.
The current study differs from the Study of Health in
Pomerania (SHIP) (16), which identified a significant association between parity and both positive TPOAb and hypoechogenic thyroid ultrasound patterns in a populationbased sample of 2156 women.
The reasons for the discrepancies between the SHIP study
(16), previous studies (14,15), and this study are not clear but
may include differences in population characteristics in terms
of age, ethnicity, iodine intake, different assays used to determine thyroid antibodies, and different cut-offs used to
define thyroid autoimmunity. The authors of the SHIP (16)
study justified their different results by arguing that the
Busselton (14) and DanThyr (15) studies were limited by relatively low response proportions of 64% and 51%, respectively, as compared with 68.8% in their study. In addition,
they believed that the exclusion of women aged 31–39 years
and 46–59 years might have led to an underestimation of the
relationship between parity and AITD in the DanThyr study
(15). They also commented that the study was performed in
an area of mild-to-moderate iodine deficiency, whereas no
exact data on iodine supply exist for the Busselton study (14).
1154
SGARBI ET AL.
Table 2. Thyroid Dysfunctions According to the Number of Parities, Abortions, and the Use of Estrogens
Parameter
No. of parities
0
1
No. of abortions
0
1
Use of estrogens
No
Yes
Euthyroidism
(n ¼ 541)
Overt
hyperthyroidism
(n ¼ 11)
Subclinical
hyperthyroidism
(n ¼ 46)
Overt
hypothyroidism
(n ¼ 7)
Hypothyroidism
(n ¼ 70)
302 (55.8)
239 (44.2)
3 (27.3)
8 (72.7)
24 (52.2)
22 (47.8)
4 (57.1)
3 (42.9)
34 (48.6)
36 (51.4)
0.31
531 (94.8)
26 (4.8)
11 (100.0)
—
43 (93.5)
3 (6.5)
6 (85.7)
1 (14.3)
65 (92.9)
5 (7.1)
0.45
487 (90.0)
54 (10.0)
10 (90.9)
1 (9.1)
42 (91.3)
4 (8.7)
7 (100.0)
—
62 (88.6)
8 (11.4)
0.97
p-Value
Data presented as no. of subjects (%).
p indicates w2 test or Fisher test.
In fact, all of these arguments could have limited the ability of
such studies (14,15) to detect an effect of parity on AITD.
However, in the present study, in which no association was
found between parity and AITD, the proportion of responders
was approximately 76% (greater than that estimated in the
SHIP study), and all women more than 30 years of age were
invited to participate in the study; additionally, the median
UIC found in our population was 210 mg/L, which is more
than optimal according to current recommendations (21). The
SHIP population was investigated during a time when the
iodine supply was shifted from deficient to a sufficient supply
(22). This is of particular importance, as previous studies have
reported a transient or persistent increase in thyroid autoantibody titers after iodine prophylaxis (23–26). Thus, it is not
possible to exclude the possibility that the parity-related risk
of AITD found in such a study (16) could have been influenced by changes in iodine supply during the study recruitment period, especially considering that the study
design did not allow the evaluation of changes in thyroid
autoimmunity by comparing the situation before and after
the iodine supplementation program (22). Further, the thyroid antibody assays and cut-offs chosen to define positive
titers of TPOAb differ between the studies. The SHIP study
differentiated between elevated (>60 mIU/mL) and positive
(>200 mIU/mL) TPOAb titers, whereas the other studies
(14,15) and our study used a TPOAb cut-off of approximately 30–35 kIU/L.
The present study also differs from a more recent casecontrol study (27), in which both female and male twins from
opposite-sex pairs had an increased frequency of thyroid antibodies compared with monozygotic pairs, indicating a potential role of microchimerism in developing thyroid
autoimmunity. However, these findings should be cautiously
interpreted, as reflected by the relatively wide confidence
intervals, and because a number of other intrauterine factors
such as number of placentas and birth weight may have
influenced the results (27).
The female preponderance in AITD might also be attributed to hormonal influences, as there is limited evidence
concerning the influence of the X chromosome in the pathogenesis of AITD (3). However, there are relatively few studies
exploring potential relationship between the use of estrogens
and thyroid autoimmunity, and the results are conflicting (3).
The present study is fully in agreement with the DanThyr
Table 3. Odds Ratios with 95% Confidence Intervals for the Association Between Positive
Thyroid Antibodies and Previous Parity, Abortion, and Use of Estrogens
Parameter
Positive TPOAb
Crude
Adjusteda
Positive TgAb
Crude
Adjusteda
Positive TPOAb and TgAb
Crude
Adjusteda
Positive TPOAb and/or TgAb
Crude
Adjusteda
a
1 Parities
(reference: no parity)
1 Abortion
(reference: no abortion)
Use of estrogens
(reference: no use)
1.21 (0.75–1.95)
1.22 (0.73–2.02)
0.73 (0.22–2.44)
0.71 (0.21–2.45)
0.61 (0.24–1.88)
0.59 (0.23–1.54)
0.95 (0.61–1.47)
1.01 (0.63–1.62)
0.81 (0.28–2.35)
0.82 (0.28–2.39)
0.85 (0.39–1.85)
0.86 (0.39–1.86)
0.95 (0.52–1.73)
0.94 (0.51–1.76)
1.28 (0.38–4.36)
0.79 (0.23–2.27)
0.84 (0.29–2.42)
1.2 (0.42–3.48)
1.11 (0.75–1.65)
0.84 (0.55–1.28)
0.63 (0.23–1.73)
0.57 (0.19–1.66)
1.39 (0.68–2.84)
0.68 (0.33–1.4)
Adjusted for age, smoking, and urinary iodine concentration.
PARITY IS NOT RELATED TO AITD
study (15), in having found no association between thyroid
autoantibodies and exogenous estrogen use.
Associations between abortions and thyroid autoimmunity
have been reported in some studies and a meta-analysis (6,28).
The reasons for these associations might include the copresence of thyroid autoimmunity with other autoimmune diseases, direct actions of thyroid antibodies on placenta, and a
greater age of women with positive thyroid antibodies and
mild thyroid failure (6,28). In fact, in the current study, women who reported at least one previous abortion had significantly higher mean serum thyroid antibodies levels
compared with those who had never experienced abortions
(Table 1). Despite this, no increased risk for thyroid autoimmunity was found in women who reported previous abortions, which is also in line with the DanThyr study (15).
In this study, age, smoking, and UIC differed significantly
between women with one or more parities and those with no
previous parity (Table 1). We do not have a reasonable explanation for age differences among the groups, but search for
a causal role of parity on thyroid autoimmunity remained
insignificant even after a logistic regression analysis adjusted
for these variables.
The major strength of our study lies in the fact that an entire
iodine-sufficient population (and not a sample) was invited to
participate, with a higher proportion of agreement than the
three previous population-based studies in this area. A
weakness of this study is the cross-section design, which is
common to all previously published studies. Also, we cannot
exclude the possibility of interaction effects across the different predictor variables, as it was not possible to do this kind of
analysis in this study. Finally, we cannot guarantee that our
findings can be generalized due to our selected population of
Japanese-Brazilians.
In summary, parity, abortion, and the use of estrogens are
not associated with AITD in the Japanese-Brazilian female
population. These findings reinforce previous reports that
argued against a key role of fetal microchimerism in the
pathogenesis of AITD.
Acknowledgments
We are grateful to the Japanese-Brazilian population and
to The Japanese-Brazilian Diabetes Study Group. The authors gratefully acknowledge Dr. Heloisa Villar for helping
in the collection of clinical and thyroid data. We thank Sirlei
Siriane for the statistical assistance. We are also grateful to
Gilberto Furuzawa and Patricia Hiroka for technical assistance and to Angela Faria for administrative assistance. This
study was supported by a grant from the São Paulo State
Research Foundation (Fundação de Amparo à Pesquisa do
Estado de São Paulo), grant 06/59737-9. R.M.B.M. is an
investigator of the Brazilian Research Council and of the
Fleury Group.
Disclosure Statement
The authors have no conflict of interest.
References
1. Wang C, Crapo L 1977 The epidemiology of thyroid disease
and implications for screening. Endocrinol Metab Clin North
Am 26:189–219.
1155
2. Tumbridge WM, Vanderpump MPJ 2000 Population
screening for autoimmune thyroid disease. Endocrinol Metab Clin North Am 29:239–253.
3. Prummel MF, Strieder T, Wiersinga WM 2004 The environment and autoimmune thyroid disease. Eur J Endocrinol
150:605–618.
4. Sgarbi JA, Maciel RMB 2009 Pathogenesis of autoimmune
thyroid diseases. Arq Bras Endocrinol Metab 53:5–14.
5. Philips DI, Lazarus JH, Butland BK 1990 The influence of
pregnancy and reproductive span on the occurrence of autoimmune thyroiditis. Clin Endocrinol (Oxf) 32:301–306.
6. Prummel MF, Wiersinga WM 2004 Thyroid autoimmunity
and miscarriage. Eur J Endocrinol 150:751–755.
7. Ando T, Davies TF 2003 Postpartum autoimmune thyroid
disease: the potential role of fetal microchimerism. J Clin
Endocrinol Metab 88:2965–2971.
8. Waldorf KMA, Nelson JL 2008 Autoimmune disease during
pregnancy and the microchimerism legacy of pregnancy.
Immunol Invest 37:631–644.
9. Koopmans M, Hovinga ICLK, Baelde HJ, Harvey MS, Heer
E, Bruijn JA, Bajema IM 2008 Chimerism occurs in thyroid,
lung, skin and lymph nodes of women with sons. J Reprod
Immunol 78:68–75.
10. Rust DW, Bianchi DW 2009 Microchimerism in endocrine
pathology. Endocr Pathol 20:11–16.
11. Ando T, Imaizumi M, Graves PN, Unger P, Davies TF 2002
Intrathyroidal fetal microchimerism in Graves’ disease. J
Clin Endocrinol Metab 87:3315–3320.
12. Renné C, Ramos Lopez E, Steimle-Grauer SA, Ziolkowski P,
Pani MA, Luther C, Holzer K, Encke A, Wahl RA, Bechstein
WO, Usadel KH, Hansmann ML, Badenhoop K 2004 Thyroid fetal male microchimerism in mothers with thyroid
disorders: presence of y-choromosomal immunofluorescence in thyroid-infiltrating lymphocytes is more prevalent
in Hashimoto’s thyroiditis and Graves’ disease than in follicular adenomas. J Clin Endocrinol Metab 89:5810–5814.
13. Klintschar M, Immel UD, Kehlen A, Schwaiger P, Mustafa T,
Mannweiler S, Regauer S, Kleiber M, Hoang-Vu C 2006 Fetal
microchimerism in Hashimoto’s thyroiditis: a quantitative
approach. Eur J Endocrinol 154:237–241.
14. Walsh JP, Bremner AP, Bulsara MK, O’Leary P, Leedman PJ,
Feddema P, Michelangeli V 2005 Parity and the risk of autoimmune thyroid disease: a community-based study. J Clin
Endocrinol Metab 90:5309–5312.
15. Bülow Pedersen I, Laurberg P, Knudsen N, Jørgensen T,
Perrild H, Ovesen L, Rasmussen LB 2006 Lack of association
between thyroid autoantibodies and parity in a population
study argues against microchimerism as a trigger of thyroid
autoimmunity. Eur J Endocrinol 154:39–45.
16. Friedrich N, Schwarz S, Thonack J, Jonh U, Wallaschofski H,
Volzke H 2008 Association between parity and autoimmune
thyroiditis in a general female population. Autoimmunity
41:174–180.
17. Gimeno SG, Ferreira SR, Franco LJ, Hirai AT, Matsumura L,
Moisés RS 2002 Prevalence and 7-year incidence of Type II
diabetes mellitus in a Japanese-Brazilian population: an
alarming public health problem. Diabetologia 45:1635–1638.
18. Sgarbi JA, Matsumura LK, Kasamatsu TS, Ferreira SR,
Maciel RMB 2010 Subclinical thyroid dysfunctions are independent risk factors for mortality in a 7.5-year follow-up:
the Japanese-Brazilian thyroid study. Eur J Endocrinol 162:
569–577.
19. Vieira JGH, Tachibana TT, Fonseca RMG, Nishida SK,
Maciel RMB 1996 Development of a immunofluorimetric
1156
20.
21.
22.
23.
24.
25.
26.
assay for the measurement of anti-thyroglobulin antibodies.
Arq Bras Endocrinol Metab 40:232–237.
Esteves RZ, Kasamatsu TS, Kunii IS, Furuzawa GK, Vieira
JGH, Maciel RMB 2007 Development of a semi-automated
method for measuring urinary iodine and its application in
epidemiological studies in Brazilian schoolchildren. Arq
Bras Endocrinol Metab 51:1477–1484.
Delange F, de Benoist B, Burgi H, the ICCIDD Working
Group 2002 Determining median urinary iodine concentration that indicates adequate iodine intake at population level. Bull World Health Organ 80:633–636.
Völzke H, Lüdemann J, Robinson DM, Spieker KW,
Schwahn C, Kramer A, John U, Meng W 2003 Prevalence of
undiagnosed thyroid disorders in a previously iodine-deficient area. Thyroid 13:803–810.
Weaver DK, Nishiyami RH, Burton WD, Batsakis JG 1966
Surgical thyroid disease: a survey before and after iodine
prophylaxis. Arch Surg 92:796–801.
Laurberg P, Pedersen B, Knudsen N, Ovesen L, Andersen S
2001 Environmental iodine intake affects the type of nonmalignant thyroid disease. Thyroid 11:457–469.
Zois C, Stavrou I, Svarna E, Seferiadis K, Tsatsoulis A 2006
Natural course of autoimmune thyroiditis after elimination
of iodine deficiency in northwestern Greece. Thyroid 16:289–
293.
Teng W, Shan Z, Teng X, Guan H, Li Y, Teng D, Jin Y, Yu X,
Fan C, Chong W, Yang F, Dai H, Yu Y, Li J, Chen Y, Zhao D,
SGARBI ET AL.
Shi X, Hu F, Mao J, Gu X, Yang R, Tong Y, Wang W, Gao T,
Li C 2006 Effect of iodine intake on thyroid diseases in
China. N Engl J Med 354:2783–2793.
27. Brix TH, Hansen PS, Kyvik KO, Hagedus L 2009 Aggregation of thyroid autoantibodies in twins from opposite-sex
pairs suggests that microchimerism may play a role in the
early stages of thyroid autoimmunity. J Clin Endocrinol
Metab 94:4439–4443.
28. Toulis KA, Goulis DG, Venetis CA, Kolibianakis EM, Negro
R, Tarlatzis BC, Papadimas I 2010 Risk of spontaneous
miscarriage in euthyroid women with thyroid autoimmunity undergoing IVF: a meta-analysis. Eur J Endocrinol 162:
643–652.
Address correspondence to:
Rui M.B. Maciel, M.D., Ph.D.
Laboratory of Molecular Endocrinology
Division of Endocrinology
Department of Medicine
Escola Paulista de Medicina
Federal University of São Paulo
Rua Pedro de Toledo 669, 11o. Andar
São Paulo 04029-032
Brazil
E-mail: [email protected]
ANEXO 3
Trabalho No. 03
Subclinical thyroid dysfunctions are independent risk factors for mortality in a
7.5-year follow-up: the Japanese-Brazilian thyroid study
José A. Sgarbi, Luiza K. Matsumura, Teresa S. Kasamatsu, Sandra R. Ferreira,
Rui M. B. Maciel
“European Journal of Endocrinology”
(Eur J Endocrinol. 2010; 162: 569 - 577)
49
European Journal of Endocrinology (2010) 162 569–577
ISSN 0804-4643
CLINICAL STUDY
Subclinical thyroid dysfunctions are independent risk factors
for mortality in a 7.5-year follow-up: the Japanese–Brazilian
thyroid study
José A Sgarbi1,2, Luiza K Matsumura1, Teresa S Kasamatsu1, Sandra R Ferreira1,3 and Rui M B Maciel1
1
Division of Endocrinology, Laboratory of Molecular Endocrinology, Department of Medicine, Escola Paulista de Medicina, Federal University of São Paulo,
Rua Pedro de Toledo 669, 11o. andar, 04029-032 São Paulo, SP, Brazil, 2Division of Endocrinology, Department of Medicine, Faculdade de Medicina de
Marı́lia, Marı́lia, Brazil and 3Department of Nutrition, School of Public Health, University of São Paulo, São Paulo, Brazil
(Correspondence should be addressed to R M B Maciel; Email: [email protected])
Abstract
Objective: The currently available data concerning the influence of subclinical thyroid disease (STD)
on morbidity and mortality are conflicting. Our objective was to investigate the relationships between
STD and cardiometabolic profile and cardiovascular disease at baseline, as well as with all-cause
and cardiovascular mortality in a 7.5-year follow-up.
Design: Prospective, observational study.
Methods: An overall of 1110 Japanese–Brazilians aged above 30 years, free of thyroid disease, and not
taking thyroid medication at baseline were studied. In a cross-sectional analysis, we investigated the
prevalence of STD and its relationship with cardiometabolic profile and cardiovascular disease.
All-cause and cardiovascular mortality rates were assessed for participants followed for up to 7.5 years.
Association between STD and mortality was drawn using multivariate analysis, adjusting for potential
confounders.
Results: A total of 913 (82.3%) participants had euthyroidism, 99 (8.7%) had subclinical
hypothyroidism, and 69 (6.2%) had subclinical hyperthyroidism. At baseline, no association was
found between STD and cardiometabolic profile or cardiovascular disease. Multivariate-adjusted
hazard ratios (HRs (95% confidence interval)) for all-cause mortality were significantly higher for
individuals with both subclinical hyperthyroidism (HR, 3.0 (1.5–5.9); nZ14) and subclinical
hypothyroidism (HR, 2.3 (1.2–4.4); nZ13) than for euthyroid subjects. Cardiovascular mortality was
significantly associated with subclinical hyperthyroidism (HR, 3.3 (1.4–7.5); nZ8), but not with
subclinical hypothyroidism (HR, 1.6 (0.6–4.2); nZ5).
Conclusion: In the Japanese–Brazilian population, subclinical hyperthyroidism is an independent risk
factor for all-cause and cardiovascular mortality, while subclinical hypothyroidism is associated with
all-cause mortality.
European Journal of Endocrinology 162 569–577
Introduction
Subclinical thyroid disease (STD) is characterized by
abnormal serum thyrotropin (TSH) levels in the
presence of free thyroxine (FT4) and total or free
triiodothyronine (FT3) within their reference ranges
(1–3). Epidemiological studies have reported a considerable prevalence of unsuspected STD in the general
population (4–6), and clinicians have more frequently
diagnosed this condition in their daily clinical practice.
The main question that a clinician faces is whether a
patient with STD requires treatment or whether an
observational strategy could be safely followed (7);
however, opinions diverge regarding the clinical significance of STD (8). Both subclinical hypothyroidism
(SChypo) (9–15) and subclinical hyperthyroidism
q 2010 European Society of Endocrinology
(SChyper) (16–20) have been associated with cardiovascular abnormalities; but there are no prospectively
validated trials, and treatment remains nonevidence
based (21–22).
One frequently raised question concerns the impact
of STD on life expectancy, but findings emerging from
epidemiological studies are very controversial on this
matter (23–29).
In this study, we estimated the prevalence of STD in
an entire Japanese–Brazilian population and assessed
its associations with cardiometabolic profile and
cardiovascular disease in individuals with unrecognized
thyroid dysfunction. We also investigated the relationship between STD at baseline and all-cause and
cardiovascular mortality in a 7.5-year follow-up.
DOI: 10.1530/EJE-09-0845
Online version via www.eje-online.org
570
J A Sgarbi and others
Methods
Study population and design
A survey was conducted in a nonmixed Japanese–
Brazilian population living in Bauru (Human Development Index 0.825; Source: www.ipeadata.gov.br), State
of São Paulo, Brazil, which aimed to estimate the
prevalence of diabetes and associated diseases in this
community. A detailed description of this survey was
reported previously (30). In summary, the entire
population of R30 years of age (nZ1751) was invited,
and 1330 (76%) individuals agreed to participate
(Fig. 1). Reasons for nonparticipation (421 individuals,
24.0%) were refusal (64.6%), change of address
(13.5%), and death (21.9%).
In the cross-sectional phase conducted in 1999–2000,
the prevalence of thyroid dysfunction and the associations of STD with cardiometabolic profile or cardiovascular disease were assessed. Individuals were followed
from 1999 to 2007 in order to investigate the influence
of STD on all-cause and cardiovascular mortality.
Study procedures
Socio-demographic, cultural, lifestyle, and health data
were obtained by standardized questionnaires and
Figure 1 Study flow diagram.
www.eje-online.org
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2010) 162
trained interviewers. A specific thyroid questionnaire
that included family and personal history of thyroid
disease was applied by experts in thyroid diseases.
Body weight and height were measured while
individuals were wearing light clothing without shoes.
Waist circumference was measured at the level of the
umbilicus while standing and during slight expiration.
Blood pressure was taken three times with an automatic
device (Omron model HEM-712C, Omron Health Care,
Bannockburn, IL, USA). The mean of the last two
measurements was used to express systolic and diastolic
blood pressure values. A standard 12-lead electrocardiogram (ECG) was obtained in the resting state by
the standard procedure and was analyzed by two
cardiologists. A Doppler probe (Imbracios 8 MHz) was
used to determine the ankle–brachial pressure index for
both extremities.
Fasting blood samples were taken and a 75-g oral
glucose tolerance test was performed. Samples were
processed for immediate analyses in the local laboratory
or were stored at K80 8C. Plasma glucose was
measured by the glucose oxidase method, while the
total cholesterol, high-density lipoprotein cholesterol
(HDL-c), and triglycerides were enzymatically evaluated
with an automatic analyzer. Low-density lipoprotein
cholesterol (LDL-c) was calculated according to the
Friedewald equation (31). Insulin concentration was
determined by a MAB-based immunofluorometric assay
(AutoDelphia, PerkinElmer Life Sciences Inc., Norton,
OH, USA). Insulin resistance was calculated by the
homeostasis model assessment (HOMA-IRZfasting
insulin (mU/ml)/22.5!fasting glycemia (mmol/l)).
Urinary iodine concentration (UIC) was measured in
early-morning urine samples by a colorimetric method
(32), with a detection limit of 10 mg/l and the normal
range between 100 and 299 mg/l.
TSH levels were measured in duplicate by a sensitive
immunofluorometric assay (Wallac–Delfia, PerkinElmer,
Turku, Finland) with a reference range of 0.45–
4.5 mU/l and functional sensitivity of 0.05 mU/l.
Serum FT4 was measured using a competitive immunoassay (Wallac–Delfia), wherein the normal reference
range was 0.7–1.5 ng/dl.
Date and cause of death were collected from death
certificates between the start of the screening
(November 1999) and June 2007. For individuals
(nZ3) who moved out of the study area and for
whom we were not able to have access to the death
certificate, we asked families about the occurrence of
death and its date and cause. In June 2007, the
ascertainment of mortality was 100%. Cardiovascular
death was defined as death from any cardiovascular or
cerebrovascular event. All-cause mortality was defined
as all deaths from any natural cause.
This study was approved by the ethics committee of
Escola Paulista de Medicina, Federal University of São
Paulo, and written informed consent was obtained from
all participants.
Subclinical thyroid dysfunction and mortality
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2010) 162
Definitions
Euthyroidism was defined as serum TSH and FT4 within
the normal reference ranges; SChyper as TSH below
0.45 mU/l with normal FT4 level; overt hyperthyroidism as TSH below 0.1 mU/l with high FT4 level; SChypo
as TSH above 4.5 mU/l with normal FT4 level; and overt
hypothyroidism as TSH above 4.5 mU/l with low FT4
level or a TSH concentration above 20 mU/l (21).
Hypertension was defined as a blood pressure
R140/90 mmHg or as the use of antihypertensive
medication; diabetes was defined according to the
American Diabetes Association criteria; and dyslipidemia was defined by the presence of any lipid
abnormality (total cholesterol levels R200 mg/dl or
triglycerides R150 mg/dl or LDL-cO130 mg/dl).
The presence of cardiovascular disease at baseline was
defined by a medical history of myocardial infarction
confirmed by a physician and by major ECG abnormalities of old infarction (Q waves) or by previous
angioplasty or any heart revascularization procedure, or
coronary insufficiency diagnosed previously by catheterization, or stroke. Peripheral arterial disease was
defined by any ankle–brachial pressure index !0.9 (33).
Statistical analysis
Prevalence rates were calculated by point and confidence interval (CI). The data were described through
absolute (n) or relative (%) frequencies, mean with S.D.,
and 95% CI. Differences in means of the baseline
characteristics according to thyroid status categories
were assessed by ANOVA (Tukey’s test for multiple
comparisons if P!0.05) or nonparametric ANOVA
(Kruskal–Wallis test), followed by the Mann–Whitney U
test. Frequencies were compared by the c2 test or the
Fisher test when one of the absolute frequencies was
below five. Variables without a normal distribution
were subjected to logarithmic transformations before
statistical analysis.
In the longitudinal analysis, survival curves according to thyroid status across the 7.5 years of follow-up
were estimated using Kaplan–Meier analysis with the
log-rank test. They were constructed considering the
death as ‘event’ and contrary cases as ‘not event’
(censorship), which were predicted by the baseline
thyroid status. Living individuals who did not complete
the 7.5 years of follow-up by June 30, 2007 were
censored for survival at 7.5 years. A Cox regression
model of proportional risks in bivariate analysis was
used to determine the crude hazard ratios (HRs).
Multivariate analysis was used to account for potential
confounders of the mortality rate. Relevant confounders
were selected by their significant association with
mortality (age, sex, presence of hypertension, diabetes
mellitus, and cardiovascular disease), which were
determined by the c2 test or the Fisher test when
frequencies were compared, and by the Student t-test or
571
the Mann–Whitney test (for variables without normal
distribution). Risk factors classically associated with
mortality were also considered (total cholesterol,
smoking status, and waist circumference), totaling a
maximum of eight risk factors (maximum of one risk
factor for every ten deaths). In cases of variables with
co-linear inter-relationships, such as diabetes and
fasting or 2-h plasma glucose levels, hypertension and
systolic blood pressure, and total cholesterol and LDL-c,
only one was considered. Models were first adjusted
for age and sex, and afterwards for the relevant
confounders. A Cox regression model of proportional
risks in bivariate analysis was used to determine multiple
HRs with 95% CI to express the adjusted relative risk of
dying for individuals classified as having STD relative
to euthyroid individuals. All statistical analyses were
performed using SAS statistical software version 9.1
(SAS Institute Inc., Cary, NC, USA). The assumed level
of significance was at P!0.05 (two-tailed).
Results
From the 1330 individuals who agreed to participate in
this cohort, we excluded those who self-reported thyroid
disease or taking thyroid medications (nZ47), and
those who reported to be using amiodarone, lithium, or
corticosteroids (nZ6). Furthermore, we excluded 167
participants for whom ECG and ankle–brachial pressure
indices were not available. Thus, 1110 individuals were
considered for the present analysis (Fig. 1). There was
no difference in demographic characteristics between
included (nZ1110) and excluded (nZ220) individuals.
Cross-sectional analysis
Prevalence rates for each thyroid status category are
presented in Table 1. The median UIC was 210 mg/l, with
no statistical difference among the thyroid status
categories (Table 2). Euthyroidism, overt, and SChyper
Table 1 Demographic characteristics and thyroid status in
Japanese–Brazilians.
Demographic characteristics
Total participants (n)
Women, n (%)
Mean age, years (S.D.)
Age distribution, n (%)
30–39 years
40–49 years
50–59 years
60–69 years
O70 years
96 (8.6)
229 (20.6)
311 (28.3)
285 (25.7)
189 (17.0)
Thyroid status, prevalence rates, % (95% CI)
Euthyroidism, nZ913
Overt hyperthyroidism, nZ20
Subclinical hyperthyroidism, nZ69
Overt hypothyroidism, nZ9
Subclinical hypothyroidism, nZ99
82.3 (80.8–84.9)
1.8 (1.0–2.6)
6.2 (4.8–7.5)
0.8 (0.3–1.3)
8.9 (7.0–10.1)
1110
591 (53.2)
56.9 (12.5)
CI, confidence interval.
www.eje-online.org
572
J A Sgarbi and others
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2010) 162
Table 2 Baseline characteristics according to thyroid status. Data are presented as meanGS.D., unless noted otherwise.
Euthyroidism
(nZ913)
Overt
hyperthyroidism
(nZ20)
Subclinical
hyperthyroidism
(nZ69)
Overt
hypothyroidism
(nZ9)
Subclinical
hypothyroidism
(nZ99)
Women, n (%)
Men, n (%)
Mean age, years (S.D.)
469 (51.4)
444 (48.6)
56.4 (12.4)
10 (50.0)
10 (50.0)
56 (12.2)
42 (60.9)
27 (39.1)
61.4 (12.5)†
7 (77.8)*
2 (22.2)
65.1 (13.4)
63 (63.6)*
36 (36.4)
58.5 (12.3)
Age distribution, n (%)
30–39 years
40–49 years
50–59 years
60–69 years
R70 years
79 (8.7)
205 (22.5)
259 (28.4)
224 (24.5)
146 (16.0)
2
4
5
7
2
5 (7.2)
8 (11.6)
13 (18.8)
24 (34.8)
19 (27.5)
–
1 (11.1)
2 (22.2)
3 (33.3)
3 (33.3)
10 (10.1)
11 (11.1)
32 (32.3)
27 (27.3)
19 (19.2)
Characteristics
BMIa (kg/m2)
Waist circumferencea (cm)
Current smoker, n (%)
Past smoker, n (%)
Hypertension, n (%)
Diabetes, n (%)
PAD, n (%)
CVD, n (%)
Statin usage, n (%)
Systolic BP (mmHg)
Diastolic BPa (mmHg)
UIC (mg/l)
TSH (mU/l)
Free T4 (ng/dl)
Fasting glucose (mg/dl)
Two-hour glucose (mg/dl)
Fasting insulin (pmol/l)
HOMA-IRa
Total cholesterol (mg/dl)
LDL-c (mg/dl)
HDL-c (mg/dl)
Triglycerides (mg/dl)
25.1 (3.9)
84.5 (10.6)
120 (13.2)
173 (19.1)
342 (37.5)
328 (35.9)
117 (12.8)
120 (13.1)
13 (1.4)
132.8G24.4
79.4G13.3
204G103
1.62G0.94
1.07G0.17
124.4G(33.5)
166.5G76.9
63.2G49.5
2.8G2.6
215.0G41.1
131.2G37.3
50.8G10.9
232.3G189.5
24.1 (2.6)
83.2 (8.8)
2 (10.0)
9 (45.0)*
6 (30.0)
11 (55.0)
3 (15.0)
5 (25.0)
–
127.5G18.7
72.4G12.8
184G99
0.1G0.11§
3.2G2.74§
137.0G(57.6)
184.3G98.8
48.8G32.3
2.5G2.1
193.2G53.7†
119.6G44.9
46.8G(9.5)
224.4G137.1
24.5 (4.2)
83.9 (9.9)
5 (7.2)
12 (17.4)
32 (46.4)
30 (43.5)
10 (14.5)
13 (18.8)
1 (1.4)
135.5G26.3
78.8G12.8
207G113
0.22G0.1§
1.12G0.18§
127.9G(40.1)
184.4G87.2
66.0G46.7
3.0G2.7
207.4G31.7
125.9G31.9
48.9G(7.2)
209.0G118.5
23.6 (2.0)
79.6 (4.9)
–
2 (22.2)
4 (44.4)
2 (22.2)
2 (22.2)
3 (33.3)
1 (11.1)*
130.1G32.4
76.2G11.4
235G68
64.7G77.3§
0.53G0.22§
113.2G(8.3)
135.3G42.4
82.6G117.7
3.2G2.7
240.8G49.9*
158.1G40.4*
55.3G(10.5)
172.8G112.9
24.5 (3.7)
82.5 (9.8)
10 (10.1)
12 (12.1)
43 (43.4)
35 (35.4)
10 (10.1)
15 (15.2)
5 (5.1)*
133.5G25.2
78.2G14.2
221G113
7.1G2.82§
1.01G0.2‡
122.6G(29.8)
161.6G87.9
63.9G55.3
2.8G2.6
214.4G47.6
126.0G43.7
50.6G(12.9)
250.7G197.9
(10.0)
(20.0)
(25.0)
(35.0)
(10.0)
BMI, body mass index; PAD, peripheral arterial disease; CVD, cardiovascular disease; BP, blood pressure; UIC, urinary iodine concentration;
TSH, thyrotropin; HOMA-IR, homeostasis model assessment for insulin resistance; LDL-c, low-density lipoprotein cholesterol; HDL-c, high-density
lipoprotein cholesterol. *P!0.05; †P!0.01; ‡P!0.001; §P!0.0001.
a
Values were log-transformed for statistical analysis.
were found in 82.3, 1.8, and 6.2% of the participants
respectively, with no significant difference in sex distribution (Table 2). On the other hand, unsuspected
overt and SChypo were identified in 0.8 and 8.9% of
the participants respectively, both significantly more
frequent in women (PZ0.04). The mean age was similar
among the groups, except for the SChyper group, in which
age was significantly higher relative to the euthyroid
group. As noted in Table 2, the expected significant
differences in TSH and FT4 levels were observed between
euthyroid individuals and those with STD.
There were no statistically significant differences
among the groups concerning body mass index, waist
circumference, smoking status, systolic or diastolic
blood pressure, fasting or 2-h plasma glucose, fasting
serum insulin, HOMA-IR, HDL-c, or triglyceride levels
(Table 2). Mean total cholesterol (PZ0.03) and LDL-c
(PZ0.02) levels were significantly increased in overt
hypothyroid subjects, but not in SChypo subjects in
comparison to euthyroid individuals. However, the
proportion of individuals undergoing statin therapy
www.eje-online.org
was significantly higher in both overt hypothyroid and
SChypo groups than in the euthyroid group (P!0.05).
The OR for statin use, adjusted for age and sex, was
significantly higher in SChypo individuals (3.4 (95% CI,
1.2–9.8)) than in the euthyroid individuals. Since statin
use could be masking a potential association between
serum levels of lipids and SChypo, the analysis was
repeated excluding this condition, but the results did
not change.
The overall proportions of diabetes, hypertension, peripheral arterial disease, and cardiovascular disease were
not statistically different among the groups (Table 2).
Longitudinal analysis
During the 7.5 years of follow-up, 83 (7.5%) deaths
were recorded in this population. Four events of death
by nonnatural causes (one by suicide and three by
trauma) were censored, and three deaths by unknown
causes were censored just for the cardiovascular death
analyses. The deaths by unknown causes occurred in
Subclinical thyroid dysfunction and mortality
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2010) 162
the euthyroid (nZ2) and in the SChyper (nZ1) group.
Deaths mainly occurred as a result of cardiovascular
causes (51.3%), cancer (22.3%), or infectious disease
(14.5%). Table 3 shows the main differences between
living and dead individuals. Among the dead subjects,
50 (65.8%) had been categorized as euthyroid, 14
(17.7%) as SChyper, and 13 (16.5%) as SChypo. No
death was notified among individuals who were
classified as having overt thyroid disease. At baseline,
serum FT4 levels were significantly higher (PZ0.018)
among dead individuals than among those who were
alive at the end of the follow-up, but no differences in
TSH levels were found between the groups.
Table 4 shows the relationship between STD and
mortality. All-cause mortality was significantly higher
in SChyper (20.3%) and SChypo (13.1%) individuals
than in euthyroid (5.7%) individuals (P!0.0001).
Kaplan–Meier analysis (Fig. 2) with the log-rank test
573
reveals higher overall mortality for both SChyper
(P!0.0001) and SChypo (PZ0.0035) groups in
comparison to the euthyroid group. Cardiovascular
mortality was significantly associated with SChyper
(P!0.0001). These differences emerged after 4 years of
the follow-up. Cox regression analysis (Table 4) revealed
that these significant associations were preserved even
after adjusting for age, sex, and multiple potential
confounders.
Discussion
In this study, we found a strong relationship between
SChyper and all-cause and cardiovascular mortality,
while SChypo was significantly associated with all-cause
mortality. These significant associations with mortality
emerged after 4 years of follow-up.
Table 3 Thyroid status, demographic characteristics, and biological variables according to vital status at the end of the follow-up. Data are
presented as meanGS.D., unless noted otherwise.
Variable
Alive (nZ1031)
Dead (nZ79)
P value
Euthyroidism
Overt hyperthyroidism
SChyper
Overt hypothyroidism
SChypo
Men, n (%)
Women, n (%)
Mean age, years
861 (83.5)
20 (1.9)
55 (5.3)
9 (0.9)
86 (8.3)
472 (45.8)
559 (54.2)
56.1G12.2
52 (65.8)
–
14 (17.7)
–
13 (16.5)
47 (59.5)
32 (40.5)
67.7G11.3
!0.0001
Distribution of age, n (%)
30–39 years
40–49 years
50–59 years
60–69 years
R70 years
Survival time, years
BMIa (kg/m2)
Waist circumferencea (cm)
Current smoker, n (%)
Past smoker, n (%)
Hypertension, n (%)
Diabetes, n (%)
PAD, n (%)
CVD, (%)
Statin usage, n (%)
Systolic BP (mmHg)
Diastolic BP (mmHg)
UIC (mg/dl)
TSH (mU/l)
Free T4 (ng/dl)
Fasting glucose (mg/dl)
Two-hour glucose (mg/dl)
Fasting insulin (pmol/l)
HOMA-IRa
Total cholesterol (mg/dl)
LDL-c (mg/dl)
HDL-c (mg/dl)
Triglycerides (mg/dl)
95 (9.2)
225 (21.8)
297 (28.8)
264 (25.6)
150 (14.5)
7.3G0.3
25.1G3.8
84.2G10.3
125 (12.2)
193 (18.8)
379 (36.8)
366 (35.5)
129 (12.5)
133 (12.9)
19 (1.8)
132.0G24.0
79.0G13.3
20.7G10.5
2.5G9.2
1.1G0.5
124.3G33.9
165.5G78.2
61.7G49.3
2.8G2.6
214.8G40.9
130.9G37.9
50.8G10.9
233.7G189.8
1 (1.3)
4 (5.1)
14 (17.7)
21 (26.6)
39 (49.4)
4.1G2.0
24.1G4.3
84.7G11.3
12 (15.2)
15 (19.0)
48 (60.8)
40 (50.6)
13 (16.5)
23 (29.1)
1 (1.3)
145.3G27.4
80.6G14.2
19.7G9.0
2.4G2.8
1.13G0.2
127.1G36.9
189.3G85.1
52.7G36.8
2.4G2.1
207.3G50.8
124.3G36.6
48.2G10.1
208.1G109.7
0.02
!0.0001
!0.0001
!0.0001
0.03
0.8
0.72
!0.0001
0.007
0.31
!0.0001
1.0
!0.0001
0.47
0.56
0.86
0.018
0.63
0.006
0.09
0.11
0.02
0.1
0.04
0.6
SChyper, subclinical hyperthyroidism; SChypo, subclinical hypothyroidism; BMI, body mass index; PAD, peripheral arterial disease; CVD, cardiovascular
disease; BP, blood pressure; UIC, urinary iodine concentration; TSH, thyrotropin; HOMA-IR, homeostasis model assessment for insulin resistance; LDL-c,
low-density lipoprotein cholesterol; HDL-c, high-density lipoprotein cholesterol.
a
Values were log-transformed for statistical analysis.
www.eje-online.org
574
J A Sgarbi and others
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2010) 162
Table 4 Hazard ratios (95% confidence interval (CI)) for 7.5-year mortality due to all and cardiovascular causes among 1110
Japanese–Brazilians.
Euthyroidism
(nZ913)
All-cause mortality, n (%)
Crude
Model 1
Model 2
Cardiovascular mortality, n (%)
Crude
Model 1
Model 2
52 (5.7)
1
1
1
26 (2.8)
1
1
1
Subclinical
hyperthyroidism
(nZ69)
Subclinical
hypothyroidism
(nZ99)
14 (20.3)
4.0 (2.2–7.2)
3.4 (1.9–6.3)
3.0 (1.5–5.9)
8 (11.6)
4.5 (2.1–10.0)
3.7 (1.6–8.4)
3.3 (1.4–7.5)
13 (13.1)
2.2 (1.2–4.3)
2.2 (1.2–4.1)
2.3 (1.2–4.4)
5 (5.1)
1.8 (0.7–4.6)
1.7 (0.6–4.3)
1.6 (0.6–4.2)
CI, confidence interval. Data are given as hazard ratio (95% CI). Model 1, adjusted for age and sex. Model 2, adjusted for Model 1 plus hypertension, diabetes
mellitus, cardiovascular disease, total cholesterol, smoking status, and waist circumference.
Despite differences in the population characteristics,
these results are similar to those found by Parle et al.
(23) and Gussekloo et al. (25), by having a significant
association between mortality and SChyper, and
because increased levels of FT4 were also associated
with increased all-cause mortality. However, we could
not confirm Gussekloo’s findings of lower mortality
among octogenarians with increased TSH levels
because our small number of individuals and events
limited our power to detect significant associations in
that age group. Such a strong association was also
reported in a recent meta-analysis (34), in which
SChyper was associated with a significant increase in
the relative likelihood of death from all causes, whereas
another meta-analysis found only a modest association
(35). On the other hand, our findings disagree with
previous studies (28, 29), which found no association
between SChyper and mortality. These studies were
larger and had a greater follow-up than the current
cohort. Therefore, both had a lower prevalence of
SChyper, and in one (28), analysis of death was based
on only three events, limiting the power to detect an
effect of SChyper on mortality.
In the present report, SCHypo was significantly
associated with death by all causes, but not with
cardiovascular mortality. However, as can be noted in
Table 4, the point estimates for association between
SChypo and cardiovascular mortality ranged from 1.6
to 1.8 in the different models, but with very large CIs.
The small number of cardiovascular deaths (five events)
probably limited our ability to detect an association
between SChypo and cardiovascular mortality, and
these HRs might have been significant with a larger
number of outcomes.
These data partially agree with a Japanese study (24),
although in such a study, the association between
SChypo and mortality disappeared by the 10-year mark.
Despite our shorter follow-up, we do not have any
evidence of a similar outcome in our population, as the
association between STD and mortality became more
pronounced throughout the study (Fig. 2). In addition,
comparisons between these cohorts should be done with
www.eje-online.org
caution, given that they have been exposed to different
iodine intake, and because such a study (24) was highly
selective in that it only included survivors of the
atomic bomb. Our findings also agree with two recent
meta-analyses (34, 35), but differ from others (36, 37)
and from recent observational studies (28, 29);
however, in one of them (29) the mean age of the
population (72.7 years) was higher than that of our
population. This difference is of particular importance
since it has been suggested (38) that SChypo is
associated with mortality in only relatively younger
populations (%65 years).
In this cohort, mortality was associated with some
metabolic and clinical variables (Table 3); however, the
causal role of STD for mortality remained significant,
even after a multivariate analysis adjusted for all
variables significantly related to mortality and for
those classically known to have an association with
mortality. Findings regarding the relationship between
STD and mortality are very discrepant, mainly because
confounders known to affect prognosis have not been
carefully considered in many studies (37).
The prevalence rates for STD in this population
confirm previous epidemiological studies reporting an
elevated prevalence of unsuspected STD in the general
population (4–6). The prevalence of SChypo in this
study was similar to that reported previously (4–6),
while the rate of SChyper was higher than that reported
for iodine-sufficient Western (4–6, 29) and Japanese
populations (39, 40). The reason for this difference is
not clear; however, similar findings were found in
another Brazilian population study (41). It has been
reported that 5 years of excessive iodine intake (1998–
2003) may have increased the prevalence of hyperthyroidism in Brazil (42), but only 17.5% of the Japanese–
Brazilians had an increased UIC, and no difference in
UIC was found among the thyroid categories. This
population could be studied during a time when iodine
supply was shifted from a mildly deficient to a sufficient
or more than sufficient supply, but unfortunately no
data for iodine status exist before the time of the study.
There is a possibility of selection bias in having included
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2010) 162
Figure 2 Kaplan–Meier survival curves for all (A) and cardiovascular (B) causes of death in Japanese–Brazilians according to
thyroid status. SChyper, subclinical hyperthyroidism; SChypo,
subclinical hypothyroidism. *(A) Log-rank test; all causes of death,
P!0.0001 for SChyper versus euthyroidism and PZ0.0035 for
SChypo versus euthyroidism. †(B) Log-rank test; cardiovascular
death, P!0.0001 for SChyper versus euthyroidism and PZ0.23 for
SChypo versus euthyroidism.
some individuals with nonthyroidal illness (43), but it is
not very likely to be of significance in this study, since
FT4 levels were significantly higher in individuals with
SChyper than in euthyroid subjects (Table 2), which is
consistent with mild thyroid hormone excess. Finally,
the use of slimming pills, a common practice in Brazil
(44), could have affected the prevalence of SChyper in
our population, but participants did not report such use.
No consistent association of STD with cardiometabolic risk factors was found at baseline in this study
Subclinical thyroid dysfunction and mortality
575
(Table 2). These findings agree with some previous large
population-based studies (4, 29, 45, 46), but differ from
others (5, 6, 28). In two of these studies (6, 28), the
difference disappeared after adjusting for other relevant
risk factors, such as age, sex, and statin use. A metaanalysis (47) found a significant decrease in total serum
cholesterol levels following L-T4 therapy, but most of the
selected studies had a nonrandomized design. In
contrast, a systematic review (48) found only marginal
evidence indicating an association between thyroid
hormone replacement and improvement in lipid profile.
Recently, a randomized, double-blind, and crossover
study of L-T4 and placebo (49) found that SChypo
treated with L-T4 improved total cholesterol and LDL-c
levels. Thus, there are no homogeneous data concerning the effects of SChypo on serum lipid levels.
We found no association of STD with cardiovascular
disease or with peripheral arterial disease at baseline in
this study, which is similar to some studies (28, 29, 50).
A modest association of SChypo with an increased risk
of coronary heart disease at baseline and at follow-up
has been found in different studies and meta-analyses
(28, 35, 36, 45, 51), but the estimated risk was close to
1.0 when only higher quality studies were pooled (35).
A recent analysis suggested that SChypo may be
associated with increased cardiovascular risk only in
middle-aged (!65 years old) individuals (38). Unfortunately, the small number of events eliminates the ability
to perform meaningful analysis according to age in the
present study.
The major strength of our study lies in our inclusion
of an entire population. In addition, participants were
examined by thyroid experts, and individuals who selfreported thyroid diseases or were taking thyroid
medications were excluded from the analysis; life status
was obtained for all participants, and mortality risk was
adjusted for multiple confounders.
This study also has several limitations, including
the fact that our data are based only on a baseline set of
thyroid tests. Thus, we cannot exclude the possibility
of influence determined by the progression from
subclinical to overt thyroid dysfunction on the risk of
mortality; however, this limitation is common to all
previously published studies. We also had a relatively
small number of cardiovascular deaths, decreasing the
power of the analysis and our ability to detect an
association with SChypo. Another limitation is the lack
of analysis stratified according to age, sex, and TSH
levels due to the small numbers of events. In addition,
we are unable to exclude the possibility of overt
thyrotoxicosis in some of our SChyper individuals,
since T3 and FT3 serum levels were not determined in
this study. Furthermore, cause of death was only
based on death certificates without additional validation by hospital records for those who died in the
hospital. Finally, we cannot guarantee the generalizability of the findings due to our selected population of
Japanese–Brazilians.
www.eje-online.org
576
J A Sgarbi and others
In summary, SChyper is an independent risk factor for
all-cause and cardiovascular mortality, whereas SChypo
is associated with increased all-cause mortality among
Japanese–Brazilians. These findings suggest that further
preventive strategies of treatment are necessary in order
to reduce mortality associated with STD in the general
population; however, to demonstrate some therapeutic
benefit, large, well-designed, randomized, and placebocontrolled trials of STD treatment will be needed. Thus,
while the treatment of STD persists as a nonevidencebased program, the choice between treating and not
treating patients with persistent endogenous STD
remains dependent on the best clinical judgment.
Declaration of interest
The authors declare that there is no conflict of interest that could be
perceived as prejudicing the impartiality of the research reported.
Funding
The Japanese–Brazilian thyroid study was supported by an unrestricted grant from the São Paulo State Research Foundation (Fundação
de Amparo à Pesquisa do Estado de São Paulo, FAPESP, grant
06/59737-9 to Dr Maciel).
Acknowledgements
We are grateful to the Japanese–Brazilian population and to the
Japanese–Brazilian Diabetes Study Group, particularly to Drs Amélia
Hirai and Sueli Gimeno. We gratefully acknowledge Dr Heloisa Villar
for helping in the collection of clinical and thyroid data. We thank
Sirlei Siani Morais for the statistical assistance. We are also grateful to
Gilberto Furuzawa and Patricia Hiroka for technical assistance and
to Angela Faria for administrative assistance. Dr Maciel is a
researcher from the Brazilian Research Council (CNPq) and from the
Fleury Institute.
References
1 Surks MI & Ocampo E. Subclinical thyroid disease. American
Journal of Medicine 1996 100 217–223.
2 Woeber KA. Subclinical thyroid dysfunction. Archives of Internal
Medicine 1997 157 1065–1068.
3 Romaldini JH, Sgarbi JA & Farah CS. Subclinical thyroid disease:
subclinical hypothyroidism and hyperthyroidism. Arquivos
Brasileiros de Endocrinologia e Metabologia 2004 48 147–158.
4 Tunbridge WM, Evered DC, Hall R, Appleton D, Brewis M, Clark F,
Evans JG, Young E, Bird T & Smith PA. The spectrum of
thyroid disease in a community: the Whickham survey. Clinical
Endocrinology 1977 7 481–493.
5 Canaris GJ, Manowitz NR, Mayor G & Ridgway C. The Colorado
thyroid disease prevalence study. Archives of Internal Medicine
2000 160 526–534.
6 Hollowell JG, Staehling NW, Flanders WD, Hannon WH,
Gunter EW, Spencer CA & Braverman LE. Serum TSH, T4,
and thyroid antibodies in the United States population (1988 to
1994): National Health and Nutrition Examination Survey
(NHANES III). Journal of Clinical Endocrinology and Metabolism
2002 87 489–499.
7 Helfand M. Screening for subclinical thyroid dysfunction in
nonpregnant adults: a summary of the evidence for the US
Preventive Service Task Force. Annals of Internal Medicine 2004
140 128–141.
www.eje-online.org
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2010) 162
8 Biondi B & Cooper DS. The clinical significance of subclinical
thyroid dysfunction. Endocrine Reviews 2008 29 76–131.
9 Biondi B, Fazio S, Palmieri EA, Carella C, Panza N, Cittadini A,
Bonè F, Lombardi G & Saccà L. Left ventricular diastolic
dysfunction in patients with subclinical hypothyroidism.
Journal of Clinical Endocrinology and Metabolism 1999 84
2064–2067.
10 Monzani F, Di Bello V, Caraccio N, Bertini A, Giorgi D, Giusti C &
Ferrannini E. Effect of levothyroxine on cardiac function and
structure in subclinical hypothyroidism: a double blind, placebocontrolled study. Journal of Clinical Endocrinology and Metabolism
2001 86 1110–1115.
11 Vitale G, Galderisi M, Lupoli GA, Celentano A, Pietropaolo I,
Parenti N, De Divitiis O & Lupoli G. Left ventricular myocardial
impairment in subclinical hypothyroidism assessed by a new
ultrasound tool: pulsed tissue Doppler. Journal of Clinical
Endocrinology and Metabolism 2002 87 4350–4355.
12 Biondi B, Palmieri EA, Lombardi G & Fazio S. Effects of subclinical
thyroid dysfunction on the heart. Annals of Internal Medicine 2002
137 904–914.
13 Aghini-Lombardi F, Di Bello V, Talini E, Di Cori A, Monzani F,
Antonangeli L, Palagi C, Caraccio N, Grazia Delle Donne M,
Nardi C, Dardano A, Balbarini A, Mariani M & Pinchera A. Early
textural and functional alterations of left ventricular myocardium
in mild hypothyroidism. European Journal of Endocrinology 2006
155 3–9.
14 Rodondi N, Bauer DC, Cappola AR, Cornuz J, Robbins J, Fried LP,
Ladenson PW, Vittinghoff E, Gottdiener JS & Newman AB.
Subclinical thyroid dysfunction, cardiac function, and the risk of
heart failure. The Cardiovascular Health study. Journal of the
American College of Cardiology 2008 52 1152–1159.
15 Akcakoyun M, Kaya H, Kargin R, Pala S, Emiroglu Y, Esen O,
Karapinar H, Kaya Z & Esen AM. Abnormal left ventricular
longitudinal functional reserve assessed by exercise pulsed wave
tissue Doppler imaging in patients with subclinical hypothyroidism. Journal of Clinical Endocrinology and Metabolism 2009 94
2979–2983.
16 Sawin CT, Geller A, Wolf PA, Belanger AJ, Baker E, Bacharach P,
Wilson PW, Benjamin EJ & D’Agostino RB. Low serum thyrotropin
concentrations as a risk factor for atrial fibrillation in older
persons. New England Journal of Medicine 1994 331 1249–1252.
17 Mercuro G, Panzuto MG, Bina A, Leo M, Cabula R, Petrini L,
Pigliaru F & Mariotti S. Cardiac function, physical exercise
capacity, and quality of life during long-term thyrotropin–
suppressive therapy with levothyroxine: effect of individual dose
tailoring. Journal of Clinical Endocrinology and Metabolism 2000 85
159–164.
18 Biondi B, Palmieri EA, Fazio S, Cosco C, Nocera M, Saccà L,
Filetti S, Lombardi G & Perticone F. Endogenous subclinical
hyperthyroidism affects quality of life and cardiac morphology and
function in young and middle-aged patients. Journal of Clinical
Endocrinology and Metabolism 2000 85 4701–4705.
19 Sgarbi JA, Villaça FG, Garbeline B, Villar HE & Romaldini JH. The
effects of early antithyroid therapy for endogenous subclinical
hyperthyroidism on clinical and heart abnormalities. Journal of
Clinical Endocrinology and Metabolism 2003 88 1672–1677.
20 Di Bello V, Aghini-Lombardi F, Monzani F, Talini E, Antonangeli L,
Palagi C, Di Cori A, Caraccio N, Delle Donne MG, Dardano A,
Pinchera A & Mariani M. Early abnormalities of left ventricular
myocardial characteristics associated with subclinical hyperthyroidism. Journal of Endocrinological Investigation 2007 30 564–571.
21 Surks MI, Ortiz E, Daniels GH, Sawin CT, Col NF, Cobin RH,
Franklyn JA, Hershman JM, Burman KD, Denke MA, Gorman C,
Cooper RS & Weissman NJ. Subclinical thyroid disease: scientific
review and guidelines for diagnosis and management. Journal of
the American Medical Association 2004 291 228–238.
22 Gharib H, Tuttle M, Baskin HJ, Fish LH, Singer PA &
McDermott MT. Subclinical thyroid dysfunction: a joint statement
on management from the American Association of Clinical
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2010) 162
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Endocrinologists, the American Thyroid Association, and the
Endocrine Society. Journal of Clinical Endocrinology and Metabolism
2005 90 581–585.
Parle JV, Maisonneuve P, Sheppard MC, Boyle P & Franklyn JA.
Prediction of all-cause and cardiovascular mortality in elderly
people from one low serum thyrotropin result: a 10-year cohort
study. Lancet 2001 358 861–865.
Imaizumi M, Akahoshi M, Ichimaru S, Nakashima E, Hida A,
Soda M, Usa T, Ashizawa K, Yokoyama N, Maeda R, Nagataki S &
Eguchi K. Risk for ischemic heart disease and all-cause mortality
in subclinical hypothyroidism. Journal of Clinical Endocrinology and
Metabolism 2004 89 3365–3370.
Gussekloo J, van Exel E, de Craen AJ, Meinders AE, Frolich M &
Westendorp RG. Thyroid status, disability and cognitive function,
and survival in old age. Journal of the American Medical Association
2004 292 2591–2599.
Rodondi N, Newman AB, Vittinghoff E, de Rekeneire N,
Satterfield S, Harris TB & Bauer DC. Subclinical hypothyroidism
and the risk of heart failure, other cardiovascular events, and
death. Archives of Internal Medicine 2005 165 2460–2466.
van den Beld AW, Visser TJ, Feelders RA, Grobbee DE &
Lamberts SW. Thyroid hormone concentrations, disease, physical
function, and mortality in elderly men. Journal of Clinical
Endocrinology and Metabolism 2005 90 6403–6409.
Walsh JP, Bremner AP, Bulsara MK, O’Leary P, Leedman PJ,
Feddema P & Michelangeli V. Subclinical thyroid dysfunction as a
risk factor for cardiovascular disease. Archives of Internal Medicine
2005 165 2467–2472.
Cappola AR, Fried LP, Arnold AM, Danese MD, Kuller LH,
Burke GL, Tracy RP & Ladenson PW. Thyroid status, cardiovascular risk, and mortality in older adults. Journal of the American
Medical Association 2006 295 1033–1041.
Gimeno SG, Ferreira SR, Franco LJ, Hirai AT, Matsumura L &
Moisés RS. Prevalence and 7-year incidence of type II diabetes
mellitus in a Japanese–Brazilian population: an alarming public
health problem. Diabetologia 2002 45 1635–1638.
Friedewald WT, Levy RJ & Fredrickson DS. Estimation of the
concentration of low-density lipoprotein cholesterol in plasma,
without use of the preparative ultracentrifuge. Clinical Chemistry
1972 18 499–502.
Esteves RZ, Kasamatsu TS, Kunii IS, Furuzawa GK, Vieira JGH &
Maciel RMB. Development of a semi-automated method for
measuring urinary iodine and its application in epidemiological
studies in Brazilian school children. Arquivos Brasileiros de
Endocrinologia e Metabologia 2007 51 1477–1484.
Sacks D, Bakal CW, Beatty PT, Becker GJ, Cardella JF, Raabe RD,
Wiener HM & Lewis CA. Position statement on the use of the
ankle–brachial index in the evaluation of patients with peripheral
vascular disease: a consensus statement developed by the
standards division of the society of cardiovascular & interventional
radiology. Journal of Vascular and Interventional Radiology 2002
13 353.
Haentjens P, Meerhaeghe AV, Poppe K & Velkeniers B. Subclinical
thyroid dysfunction and mortality: an estimate of relative and
absolute excess all-cause mortality based on time-to-event data
from cohort studies. European Journal of Endocrinology 2008 159
329–341.
Ochs N, Auer R, Bauer DC, Nanchen D, Gussekloo J, Cornuz J &
Rodondi N. Meta-analysis: subclinical thyroid dysfunction and the
risk for coronary heart disease and mortality. Annals of Internal
Medicine 2008 148 832–845.
Rodondi N, Aujesky D, Vittinghoff E, Cornuz J & Bauer DC.
Subclinical hypothyroidism and the risk of coronary heart disease:
a meta-analysis. American Journal of Medicine 2006 119 541–551.
Volzke H, Schwahn C, Wallaschofski H & Dorr M. Review: the
association of thyroid dysfunction with all-cause and circulatory
mortality: is there a causal relationship? Journal of Clinical
Endocrinology and Metabolism 2007 92 2421–2429.
Subclinical thyroid dysfunction and mortality
577
38 Razvi S, Shakoor A, Vanderpump M, Weaver JU & Pearce SH. The
influence of age on the relationship between subclinical
hypothyroidism and ischemic heart disease: a metaanalysis.
Journal of Clinical Endocrinology and Metabolism 2008 93
2998–3007.
39 Okamura K, Nakashima T, Ueda K, Inoue K, Omae T &
Fujishima M. Thyroid disorders in the general population of
Hisayama Japan, with special reference to prevalence and
sex differences. International Journal of Epidemiology 1987 16
545–549.
40 Konno N, Yuri K, Taguchi H, Miura K, Taguchi S, Hagiwara K &
Murakami S. Screening for thyroid diseases in an iodine
sufficient area with sensitive thyrotrophin assays, and serum
thyroid autoantibody and urinary iodine determinations. Clinical
Endocrinology 1993 38 273–281.
41 Sichieri R, Baima J, Henriques J, Vasconcellos M, Marante T,
Kumagai S & Vaisman M. Prevalence of thyroid disease and
positive antitiroperoxidase among 1,500 women 35 year old and
older: a population-based survey in the city of Rio de Janeiro,
Brazil. Thyroid 2005 15 S-42 (abs 118).
42 Camargo RY, Tomimori EK, Neves SC, Rubio IGS, Galrão AL,
Knobel M & Medeiros-Neto G. Thyroid and the environment:
exposure to excessive nutritional iodine increases the prevalence
of thyroid disorders in Sao Paulo, Brazil. European Journal of
Endocrinology 2008 159 293–299.
43 Adler SM & Wartofsky L. The nonthyroidal illness syndrome.
Endocrinology and Metabolism Clinics of North America 2007 36
657–672.
44 Sichieri R, Andrade R, Baima J, Henriques J & Vaisman M. TSH
levels associated with slimming pill use in a population-based
study of Brazilian women. Arquivos Brasileiros de Endocrinologia e
Metabologia 2007 51 1448–1451.
45 Hak AE, Pols HA, Visser TJ, Drexhage HA, Hofman A &
Witteman JC. Subclinical hypothyroidism is an independent risk
factor for atherosclerosis and myocardial infarction in elderly
women: the Rotterdam Study. Annals of Internal Medicine 2000
132 270–278.
46 Takashima N, Niwa Y, Mannami T, Tomoike H & Iwai N.
Characterization of subclinical thyroid dysfunction from cardiovascular and metabolic viewpoints. The Suita Study. Circulation
Journal 2007 71 191–195.
47 Danese MD, Ladenson PW, Meinert CL & Powe NR. Effect of
thyroxine therapy on serum lipoproteins in patients with mild
thyroid failure: a quantitative review of the literature. Journal of
Clinical Endocrinology and Metabolism 2000 85 2993–3001.
48 Villar HCCE, Saconato H, Valente O, Atallah AN. Thyroid hormone
replacement for subclinical hypothyroidism (Cochrane Review
CD003419). In: The Cochrane Library, Issue 3, 2007.
49 Razvi S, Ingoe L, Keeka G, Oates C, McMillan C & Weaver JU.
The beneficial effect of L-thyroxine on cardiovascular risk
factors, endothelial function, and quality of life in subclinical
hypothyroidism: randomized, crossover trial. Journal of Clinical
Endocrinology and Metabolism 2007 92 1715–1723.
50 Vanderpump MP, Tunbridge WM, French JM, Appleton B, Bates D,
Clark F, Grimley Evans J, Rodgers H, Tunbridge F & Young ET. The
development of ischemic heart disease in relation to autoimmune
thyroid disease in a 20-year follow-up study of an English
community. Thyroid 1996 6 155–160.
51 Singh S, Duggal J, Molnar J, Maldonado F, Barsano CP & Arora R.
Impact of subclinical thyroid disorders on coronary heart
disease, cardiovascular and all-cause mortality: a meta-analysis.
International Journal of Cardiology 2008 125 41–48.
Received 26 November 2009
Accepted 2 December 2009
www.eje-online.org
ANEXO 4
Trabalho No. 04
Subclinical hypothyroidism and the risk of coronary heart disease and mortality
Nicolas Rodondi, Wendy P. J. den Elzen, Douglas C. Bauer, Anne R. Cappola, Salman
Razvi, John P. Walsh, Bjørn O. Åsvold, Giorgio Iervasi, Misa Imaizumi, Team H. Collet,
Alexandra Bremner, Patrick Maisonneuve, José A. Sgarbi, Khaw KT, Mark
Vanderpump, Anne B. Newman, Jacques Cornuz, Jayne A. Franklyn, Westendorp RG,
Eric Vittinghoff, Jacobijn Gussekloo; for the Thyroid Studies Collaboration
“Journal of the American Medical Association”
(JAMA 2010; 304:1365-1374)
59
REVIEW
CLINICIAN’S CORNER
Subclinical Hypothyroidism and the Risk
of Coronary Heart Disease and Mortality
Nicolas Rodondi, MD, MAS
Wendy P. J. den Elzen, MSc
Douglas C. Bauer, MD
Anne R. Cappola, MD, ScM
Salman Razvi, MD, FRCP
John P. Walsh, MBBS, FRACP, PhD
Bjørn O. Åsvold, MD, PhD
Giorgio Iervasi, MD
Misa Imaizumi, MD, PhD
Tinh-Hai Collet, MD
Alexandra Bremner, PhD
Patrick Maisonneuve, Ing
José A. Sgarbi, MD
Kay-Tee Khaw, MD
Mark P. J. Vanderpump, MD, FRCP
Anne B. Newman, MD, MPH
Jacques Cornuz, MD, MPH
Jayne A. Franklyn, MD, PhD, FRCP
Rudi G. J. Westendorp, MD, PhD
Eric Vittinghoff, PhD
Jacobijn Gussekloo, MD, PhD
for the Thyroid Studies Collaboration
C
ONTROVERSY PERSISTS ON THE
indications for screening and
threshold levels of thyroidstimulating hormone (TSH)
for treatment of subclinical hypothyroidism,1-3 defined as elevated serum TSH
levels with normal thyroxine (T4) concentrations. Because subclinical hypothyroidism has been associated with hypercholesterolemia4 and atherosclerosis,5
See also Patient Page.
CME available online at
www.jamaarchivescme.com
and questions on p 1392.
Context Data regarding the association between subclinical hypothyroidism and cardiovascular disease outcomes are conflicting among large prospective cohort studies.
This might reflect differences in participants’ age, sex, thyroid-stimulating hormone
(TSH) levels, or preexisting cardiovascular disease.
Objective To assess the risks of coronary heart disease (CHD) and total mortality
for adults with subclinical hypothyroidism.
Data Sources and Study Selection The databases of MEDLINE and EMBASE (1950
to May 31, 2010) were searched without language restrictions for prospective cohort
studies with baseline thyroid function and subsequent CHD events, CHD mortality,
and total mortality. The reference lists of retrieved articles also were searched.
Data Extraction Individual data on 55 287 participants with 542 494 person-years
of follow-up between 1972 and 2007 were supplied from 11 prospective cohorts in
the United States, Europe, Australia, Brazil, and Japan. The risk of CHD events was
examined in 25 977 participants from 7 cohorts with available data. Euthyroidism was
defined as a TSH level of 0.50 to 4.49 mIU/L. Subclinical hypothyroidism was defined
as a TSH level of 4.5 to 19.9 mIU/L with normal thyroxine concentrations.
Results Among 55 287 adults, 3450 had subclinical hypothyroidism (6.2%) and 51 837
had euthyroidism. During follow-up, 9664 participants died (2168 of CHD), and 4470
participants had CHD events (among 7 studies). The risk of CHD events and CHD mortality increased with higher TSH concentrations. In age- and sex-adjusted analyses,
the hazard ratio (HR) for CHD events was 1.00 (95% confidence interval [CI], 0.861.18) for a TSH level of 4.5 to 6.9 mIU/L (20.3 vs 20.3/1000 person-years for participants with euthyroidism), 1.17 (95% CI, 0.96-1.43) for a TSH level of 7.0 to 9.9
mIU/L (23.8/1000 person-years), and 1.89 (95% CI, 1.28-2.80) for a TSH level of 10
to 19.9 mIU/L (n=70 events/235; 38.4/1000 person-years; P⬍.001 for trend). The
corresponding HRs for CHD mortality were 1.09 (95% CI, 0.91-1.30; 5.3 vs 4.9/
1000 person-years for participants with euthyroidism), 1.42 (95% CI, 1.03-1.95; 6.9/
1000 person-years), and 1.58 (95% CI, 1.10-2.27, n=28 deaths/333; 7.7/1000 personyears; P=.005 for trend). Total mortality was not increased among participants with
subclinical hypothyroidism. Results were similar after further adjustment for traditional cardiovascular risk factors. Risks did not significantly differ by age, sex, or preexisting cardiovascular disease.
Conclusions Subclinical hypothyroidism is associated with an increased risk of CHD
events and CHD mortality in those with higher TSH levels, particularly in those with a
TSH concentration of 10 mIU/L or greater.
www.jama.com
JAMA. 2010;304(12):1365-1374
screening and treatment have been advocated to prevent cardiovascular disease.3 However, data on the associations with coronary heart disease (CHD)
events and mortality are conflicting
among several large prospective cohorts.6-9 Three recent study-level metaanalyses10-12 found modestly increased
©2010 American Medical Association. All rights reserved.
risks for CHD and mortality, but with
heterogeneity among individual studies that used different TSH cutoffs, difAuthor Affiliations are listed at the end of this article.
Corresponding Author: Nicolas Rodondi, MD, MAS, Department of Ambulatory Care and Community Medicine,
University of Lausanne, Bugnon 44, 1011 Lausanne,
Switzerland ([email protected]).
(Reprinted) JAMA, September 22/29, 2010—Vol 304, No. 12
Downloaded from jama.ama-assn.org by guest on 14 December 2010
1365
SUBCLINICAL HYPOTHYROIDISM AND RISK OF CORONARY HEART DISEASE, MORTALITY
ferent confounding factors for adjustment, and varying CHD definitions.10
Part of the heterogeneity might also be
related to differences in participants’ age,
sex, or severity of subclinical hypothyroidism (as measured by TSH level).4
One cohort study suggested particularly high risk in participants with subclinical hypothyroidism and preexisting cardiovascular disease.8
Analysis of individual participant
data from large cohort studies may reconcile these conflicting data and define the influence of age, TSH levels, and
preexisting cardiovascular disease. Individual participant data analysis is considered the best way for synthesizing
evidence across several studies because it is not subject to potential bias
from study-level meta-analyses (ecological fallacy)13 and allows performance of time-to-event analyses.14
To clarify the cardiovascular risk of
subclinical hypothyroidism, we formed
the Thyroid Studies Collaboration and
conducted an individual participant
data analysis of subclinical hypothyroidism and CHD outcomes.
METHODS
Identification of potential studies was
based on protocols developed for our
study-level meta-analysis of prospective cohort studies.10 Briefly, we conducted a systematic literature search of
articles in all languages on the association between subclinical thyroid dysfunction and CHD or mortality (cardiovascular and total) published from
1950 to May 31, 2010, in the MEDLINE
and EMBASE databases and searched
bibliographies of key articles (details are
available in the eMethods at http://www
.jama.com). To maximize the quality
and comparability of the studies, we formulated general inclusion criteria a
priori. We included only full-text, published longitudinal cohort studies that
(1) measured thyroid function with
both serum TSH level and thyroxine
(T4) level at baseline in adults, (2) followed up participants systematically
over time, (3) assessed CHD events
and/or mortality, and (4) had a comparison group with euthyroidism. We
excluded studies that only examined
persons taking antithyroid medications, thyroxine replacement or amiodarone, or with overt hypothyroidism
(defined as low T4 and elevated TSH
concentrations). Possible studies for inclusion were independently assessed for
suitability by 2 of the authors (N.R.,
J.G.) and any disagreement was resolved by discussion with a third author (D.C.B). The agreement between
the 2 reviewers was 99.9% for the first
screen (titles and abstracts, ␬=0.98) and
100% for the full-text screen (␬=1.00).
Investigators from each eligible study
were invited to join the Thyroid Studies Collaboration. We collected detailed
informationaboutprespecifiedoutcomes
and potential confounding variables for
each participant. Requested data included individual demographic characteristics, baseline thyroid function (TSH
and T4 levels), baseline cardiovascular
risk factors (eg, low- and high-density
lipoprotein cholesterol level, diabetes,
blood pressure, cigarette smoking),
prevalent cardiovascular disease, medication use at baseline (thyroid medication, lipid-lowering and antihypertensive drugs), and outcome data.
To maximize the comparability of the
studies, we used a common definition of
subclinical hypothyroidism. Based on expert reviews1,2 and definitions used in the
Cardiovascular Health Study,6,15 we defined subclinical hypothyroidism as a serum TSH level of 4.5 mIU/L or greater to
less than 20 mIU/L, with a normal T4 concentration;andeuthyroidismwasdefined
as a serum TSH level of 0.5 mIU/L or
greater and less than 4.5 mIU/L. Because
the Whickham Survey used a firstgeneration TSH radioimmunoassay,
which gives higher measured TSH values than current assays,16 a TSH range of
6.0 mIU/L or greater to less than 21.5
mIU/L was used for this individual participant data analyses, as in the original
and recent analysis of this study.17,18 In
thatstudy,aserumTSHlevelof6.0mIU/L
corresponded to the 97.5th percentile of
the group with negative thyroid antibodies,18 which is close to the modern level
of 4.5 mIU/L for the current generation
of assays. For T4 level, we used site- and
1366 JAMA, September 22/29, 2010—Vol 304, No. 12 (Reprinted)
method-dependent specific cutoffs
(eTable at http://www.jama.com) because T4 measurements show greater intermethod variation than do sensitive
TSH assays. The Whickham Survey measured total T4 level.18 Participants with
abnormal T4 values, results suggestive
of nonthyroidal illness (low TSH and FT4
levels) or low TSH level (⬍0.5 mIU/L)
were excluded (n=3023). Some studies had participants with missing T4 values (eTable); we considered participants with a TSH level of 4.5 mIU/L to
19.9 mIU/L and a missing T4 level as having subclinical hypothyroidism because most adults with this degree of
TSH elevation have subclinical and not
overt hypothyroidism. 1 9 We performed a sensitivity analysis excluding
those with a missing T4 level.
Outcome measures were CHD
events, CHD mortality, and total mortality. To limit outcome heterogeneity
observed with previous study-level
meta-analyses,10-12 we used more homogeneous outcome definitions. Similar to the current Framingham risk
score,20 we limited cardiovascular mortality to CHD mortality or sudden death
(eTable). A CHD event was defined as
nonfatal myocardial infarction or CHD
death (equivalent to hard events in the
Framingham risk score20) and hospitalization for angina or coronary revascularization (coronary artery bypass
grafting or angioplasty). 6 We performed a sensitivity analysis with hard
events only.
Using previously described criteria10
and new information from study authors, we systematically evaluated the following key indicators of study quality13: methods of outcome adjudication
and ascertainment, accounting for confounders, and completeness of follow-up ascertainment. Two reviewers
(N.R., J.G.) rated all studies for quality.
We used separate Cox proportional
hazard models to assess the associations of subclinical hypothyroidism with
CHD events and mortality for each cohort (SAS version 9.2, SAS Institute Inc,
Cary, North Carolina). Pooled estimates for each outcome were calculated using random-effects models, based
©2010 American Medical Association. All rights reserved.
Downloaded from jama.ama-assn.org by guest on 14 December 2010
SUBCLINICAL HYPOTHYROIDISM AND RISK OF CORONARY HEART DISEASE, MORTALITY
on the variance model according to
DerSimonian and Laird,21 as recommended14,22 and published in recent
2-stage individual participant data analyses.23 Results were summarized using forest plots (Review Manager version 5.0.24,
Nordic Cochrane Centre, Copenhagen,
Denmark). The research authors of
1 study with 14 CHD outcomes 5,10
declined to participate; as recommended,24 we included the published
summary estimate from that study in the
random-effects models in a sensitivity
analysis. To assess heterogeneity across
studies, we used the I2 statistic, which describes the total variation across studies
attributable to heterogeneity rather than
chance (I2⬎50% indicating at least moderate statistical heterogeneity).25
Primary analyses were adjusted for
age and sex, and then for traditional cardiovascular risk factors (systolic blood
pressure, smoking, total cholesterol,
diabetes) that were available in all cohorts (except for the Birmingham
Study,26 which was excluded from this
analysis). We considered the age- and
sex-adjusted model as the primary
analysis because some traditional risk
factors are potential mediators of the relationship between subclinical hypothyroidism and CHD.4
To explore sources of heterogeneity,
we performed several predefined subgroup and sensitivity analyses. We conducted stratified analyses by age, sex, race,
TSH concentrations, and preexisting cardiovascular disease. Based on expert
reviews1,2 and previous studies,7,15 subclinical hypothyroidism was stratified
according to the following TSH concentration categories: 4.5-6.9 mIU/L (mild
elevation), 7.0-9.9 mIU/L (moderate
elevation), and 10.0-19.9 mIU/L (marked
elevation). In the study-specific analyses stratified by age or TSH level, some
strata had participants without either
CHD deaths or CHD events (for 1
study27). For these study-specific analyses, we used penalized likelihood methods28 to obtain hazard ratios (HRs) and
confidence intervals (CIs). As done in
previous studies,7,27,29 after including all
participants in the primary analyses, we
performed sensitivity analyses exclud-
ing participants who had thyroid hormone use at baseline and during followup. To calculate age- and sex-adjusted
rates per 1000 person-years, we first fit
Poisson models30 to the pooled data, then
standardized the fitted rate in the euthyroidism group to the overall age and sex
distribution of the pooled sample. Finally,
to obtain rates in the TSH groups consistent with the meta-analytic results, we
multiplied the standardized rates in the
euthyroidism group by the summary
meta-analytic HRs. We checked the proportional hazard assumption using
graphical methods and Schoenfeld tests
(all P⬎.05). We used the Egger test31 and
age- and sex-adjusted funnel plots to
assess for publication bias.
RESULTS
Among 4440 reports identified, 12 prospective studies met eligibility criteria
(eFigure at http://www.jama.com) and
11 prospective cohorts in the United
States, Europe, Australia, Brazil, and Japan agreed to provide individual participant data (TABLE 1). The final sample included 55 287 adults comprising 3450
with subclinical hypothyroidism (6.2%)
and 51 837 with euthyroidism. Zero to
8.3% of participants reported thyroid
hormone use at baseline (all excluded in
5 studies) and 0% to 12.6% reported thyroid hormone use during follow-up. The
median follow-up ranged from 2.5 to 20
years, with total follow-up of 542 494
person-years.
All 11 cohort studies reported total
and CHD deaths, and 7 studies also reported CHD events among 25 977 participants. For the quality of individual
studies, all studies reported outcome adjudication without knowledge of thyroid status; 4 of 7 studies reporting CHD
events used formal adjudication procedures6-8,27; and 4 of 11 studies reporting CHD deaths mainly used death certificates.26,33-35 All studies had 5% or less
loss to follow-up.
During follow-up, 9664 participants
died (2168 of CHD) and 4470 participants had CHD events (among 7 studies). In age- and sex-adjusted analyses,
the overall HR for participants with subclinical hypothyroidism compared with
©2010 American Medical Association. All rights reserved.
euthyroidism was 1.18 (95% CI, 0.991.42) for CHD events (24.0 vs 20.3/
1000 person-years for participants with
euthyroidism), 1.14 (95% CI, 0.991.32) for CHD mortality (5.5 vs 4.9/
1000 person-years), and 1.09 (95% CI,
0.96-1.24) for total mortality (23.1 vs
21.1/1000 person-years; FIGURE 1). We
found heterogeneity across studies for
CHD events (I2 =59%) and total mortality (I2 =66%), but not for CHD mortality (I2 =0%). We subsequently examined whether heterogeneity was related
to differences in risks by degree of subclinical hypothyroidism and age. The
risk of CHD events (P⬍.001 for trend)
and CHD death (P =.005 for trend) increased with higher TSH level, but not
for total mortality (FIGURE 2). In stratified analyses, participants with TSH levels of 10 mIU/L or greater had significantly increased risk of CHD events
(HR, 1.89 [95% CI, 1.28-2.80]; n=70
events/235; 38.4 vs 20.3/1000 personyears for participants with euthyroidism) and CHD mortality (HR, 1.58
[95% CI, 1.10-2.27]; n = 28 deaths/
333; 7.7 vs 4.9/1000 person-years) compared with participants with euthyroidism. The risk for CHD associated with
subclinical hypothyroidism appeared to
be somewhat higher in younger participants, but the number of outcomes
in the younger age group was small, and
there was no significant trend in CHD
risk across age groups. Otherwise, the
risk estimates for CHD events, CHD
mortality, and total mortality did not
differ significantly according to age, sex,
race, or preexisting cardiovascular disease, except an increase in CHD events
and CHD mortality among white but
not among nonwhite participants with
subclinical hypothyroidism (TABLE 2).
All results were similar after further adjustment for traditional cardiovascular risk factors.
Sensitivity analyses yielded similar results, with increased risks of CHD
events and mortality in those with TSH
levels of 10 mIU/L or greater (TABLE 3).
Risk estimates were slightly higher for
those with TSH levels of 10 mIU/L or
greater after excluding those who took
thyroid medication during follow-up.
(Reprinted) JAMA, September 22/29, 2010—Vol 304, No. 12
Downloaded from jama.ama-assn.org by guest on 14 December 2010
1367
SUBCLINICAL HYPOTHYROIDISM AND RISK OF CORONARY HEART DISEASE, MORTALITY
COMMENT
In this analysis of 55 287 individual participants from 11 prospective cohort
studies, subclinical hypothyroidism was
associated with an increased risk of
CHD events and CHD mortality in
those with higher TSH levels. There was
a significant trend of increased risk at
higher serum TSH concentrations, and
the risk of both CHD mortality and
CHD events was significantly increased in participants with TSH levels of 10 mIU/L or greater. These associations persisted after adjustment for
traditional cardiovascular risk factors,
and did not significantly differ by age,
Estimates were lower for subclinical hypothyroidism overall after limiting the
analyses to 4 studies with formal adjudication procedures, but slightly higher
for those with TSH levels of 10 mIU/L
or greater. The effect of increasing TSH
level on CHD events did not significantly differ according to age (P=.87 for
interaction). We found no evidence of
publication bias, either with visual assessment of age- and sex-adjusted funnel plots or with the Egger test for mortality data (P = .39 for CHD mortality
and P=.97 for total mortality) and little
evidence of publication bias for CHD
events (P=.13 for CHD events).
sex, race, or preexisting cardiovascular disease. Compared with participants with euthyroidism, the overall HR
for CHD events with subclinical hypothyroidism was 1.18 (95% CI, 0.991.42) and the overall HR for CHD mortality was 1.14 (95% CI, 0.99-1.32).
Minimal TSH elevations were not associated with an increased risk of CHD
events and CHD mortality. Our results clarify the CHD risk of subgroups of adults with subclinical hypothyroidism, which could not be
adequately addressed in previous studylevel meta-analyses10-12 or in single cohort studies performed in more lim-
Table 1. Baseline Characteristics of Individuals in Included Studies (N=55 287)
Thyroid Medication
Use, No. (%)
No. (%)
Study
Description of
Study Sample
No.
Age,
Median
(Range), y a
Cardiovascular
Health
Study,6 2006
CDAs with Medicare
eligibility in 4 US
communities
3003
71
(64-100)
Health, Aging,
and Body
Composition
Study,7 2005
CDAs aged 70-79 y with
Medicare eligibility in 2
US communities
2660
74
(69-81)
1098
68
(60-94)
12 617
24 590
Birmingham
CDAs aged ⱖ60 y from
Study,26 2001
primary care practice in
Birmingham, England
EPIC-Norfolk
Adults aged 45-79 y living
Study,32 2010
in Norfolk, England
HUNT Study,33
2008
Adults aged ⬎40 y living in
Nord-Trøndelag
County, Norway
All adults aged 85 y living in
Leiden 85-plus
Leiden, the Netherlands
Study,27 2004
8
Women
Subclinical
At
During
Hypothyroidism Baseline c Follow-up
United States
1803 (60.0)
492 (16.4)
Follow-up b
Start, y
Duration,
Median Person(IQR), y
Years
0
153 (5.1)
1989-1990
13.9
(8.7-16.4)
36 865
335 (12.6)
222 (8.3)
334 (12.6)
1997
8.3
(7.3-8.4)
19 410
Europe
622 (56.6)
92 (8.4)
0
28 (2.6)
1988
10.2
(5.9-10.6)
9030
58
(39-78)
6828 (54.1)
720 (5.7)
0
NA
1995-1998
55
(41-98)
16 744 (68.1)
814 (3.3)
0
NA
1995-1997
8.3
(7.9-8.9)
200 334
16 (3.3)
1997-1999
5.2
(2.5-8.5)
2624
2000-2006
2.5
(1.6-3.7)
7710
1338 (50.3)
486
85
(NA)
318 (65.4)
35 (7.2)
14 (2.9)
12.7
153 845
(12.0-13.6)
Pisa cohort,
2007
Patients admitted to
cardiology department
in Pisa, Italy d
2875
63
(19-92)
921 (32.0)
228 (7.9)
12 (0.4)
Whickham
Survey,17,18
1996, 2010
Adults living in and near
Newcastle upon Tyne,
England
2406
46
(18-92)
1284 (53.4)
124 (5.2)
99 (4.1)
73 (3.0)
1972-1974
19
(15-20)
39 084
Busselton Health
Study,9 2005
Adults living in Busselton,
Western Australia
1984
51
(18-90)
89 (4.5)
15 (0.8)
33 (1.7)
1981
20.0
(19.4-20.0)
35 158
Atomic bomb survivors in
Nagasaki Adult
Nagasaki, Japan
Health
Study,34 2004
2591
57
(38-92)
420 (16.2)
33 (1.3)
6 (0.2)
1984-1987
13.1
(12.3-13.7)
31 559
977
56
(30-92)
NA
1999-2000
7.3
(7.0-7.5)
Brazilian Thyroid Adults of Japanese descent
Study,35 2010
living in São Paulo,
Brazil
Australia
973 (49.0)
Asia
1586 (61.2)
South America
518 (53.0)
101 (10.3)
0
0
6875
Abbreviations: CDA, community-dwelling adult; IQR, interquartile range (25th-75th percentiles); NA, data not available.
a Participants younger than 18 years were not included.
b For all cohorts, the maximal follow-up data that were available were used, which might differ from previous reports for some cohorts.
c The numbers of participants with thyroid medication use and thyroid-stimulating hormone levels of 10 mIU/L or greater were 12 of 222 in the Health, Aging, and Body Composition
Study; 3 of 14 in the Leiden 85-plus Study; 12 of 12 in the Pisa cohort; 2 of 99 in the Whickham Survey; 2 of 15 in the Busselton Health Study; and 2 of 33 in the Nagasaki Adult
Health Study.
d Excluded patients with acute coronary syndrome or severe illness.
1368 JAMA, September 22/29, 2010—Vol 304, No. 12 (Reprinted)
©2010 American Medical Association. All rights reserved.
Downloaded from jama.ama-assn.org by guest on 14 December 2010
SUBCLINICAL HYPOTHYROIDISM AND RISK OF CORONARY HEART DISEASE, MORTALITY
participant characteristics (eg, age, TSH
concentrations) because of potential
bias without individual participant data
analysis (ecological fallacy),13 and they
also were limited by clinical heterogeneity,10,36 with individual studies using
varying TSH cutoffs, confounding factors for adjustment, and CHD definitions. Among 11 cohorts, only 2 studies previously reported results stratified
ited age groups or without TSH
stratification.6,7,26,27
These results are generally consistent with previous study-level metaanalyses showing modest increased
risks of CHD events and cardiovascular mortality associated with subclinical hypothyroidism.10,11 However, these
meta-analyses could not accurately explore potential differences related to
by TSH level. One study9 reported an
increased risk of CHD events in participants with a TSH level of 10.0 mIU/L
or greater (HR, 2.2; 95% CI, 1.2-4.2)
and the other study7 reported an increased risk of cardiovascular mortality (HR, 2.26; 95% CI, 0.54-9.45) but
not CHD events (HR, 0.96; 95% CI,
0.35-2.61) over 4 years among adults
aged 70 to 79 years with TSH levels of
Figure 1. Subclinical Hypothyroidism vs Euthyroidism for Coronary Heart Disease (CHD) Events, CHD Mortality, and Total Mortality a
Subclinical
Hypothyroidism
No. of
Events
180
CHD Eventsb
Cardiovascular Health Study,6 2006
Health, Aging, and Body Composition Study,7 2005
EPIC-Norfolk Study,32 2010
Leiden 85-plus Study,27 2004
No. of
Participants
492
Euthyroidism
No. of
Events
955
No. of
Participants
2511
HR
(95% CI)
1.00 (0.85-1.17)
62
335
493
2325
0.89 (0.68-1.16)
17.4
103
720
1575
11 897
1.09 (0.89-1.33)
20.8
4.5
7
35
76
449
1.29 (0.59-2.80)
Pisa cohort,8 2007
20
228
148
2647
1.72 (1.07-2.74)
9.7
Whickham Survey,17,18 1996, 2010
27
121
438
2239
1.32 (0.89-1.96)
11.9
Busselton Health Study,9 2005
31
89
355
1889
1.78 (1.22-2.58)
12.8
430
2020
4040
23 957
1.18 (0.99-1.42)
100.0
Total (I 2 = 59%)
Decreased
Risk
Weight,
%
22.9
0.2
0.5
Increased
Risk
1
2
5
2
5
2
5
HR (95% CI)
CHD Mortalityc
Cardiovascular Health Study,6 2006
75
491
365
2511
1.09 (0.85-1.40)
Health, Aging, and Body Composition Study,7 2005
19
335
156
2325
0.85 (0.53-1.37)
33.8
9.2
Birmingham Study,26 2001
11
92
113
1006
1.21 (0.64-2.29)
5.2
EPIC-Norfolk Study,32 2010
31
720
422
11 897
1.19 (0.83-1.72)
15.6
HUNT Study,33 2008
24
814
375
23 776
1.09 (0.72-1.65)
12.1
3
35
41
446
0.87 (0.27-2.82)
1.5
Pisa cohort,8 2007
14
228
92
2647
1.91 (1.08-3.36)
6.6
Whickham Survey,17,18 1996, 2010
16
124
223
2282
1.08 (0.64-1.81)
7.8
Busselton Health Study,9 2005
13
89
144
1892
1.67 (0.94-2.97)
6.3
Leiden 85-plus Study,27 2004
Nagasaki Adult Health Study,34 2004
Total (I 2 = 0%)
4
420
27
2171
0.67 (0.23-1.91)
1.9
210
3348
1958
50 953
1.14 (0.99-1.32)
100.0
0.2
0.5
1
HR (95% CI)
Total Mortality
Cardiovascular Health Study,6 2006
Health, Aging, and Body Composition Study,7 2005
310
492
1514
2511
1.07 (0.95-1.21)
13.8
92
335
699
2325
0.89 (0.72-1.11)
11.0
Birmingham Study,26 2001
32
92
435
1006
0.92 (0.64-1.33)
7.1
EPIC-Norfolk Study,32 2010
108
720
1716
11 897
0.97 (0.80-1.18)
11.6
HUNT Study,33 2008
116
814
2159
23 776
0.99 (0.82-1.19)
11.9
Leiden 85-plus Study,27 2004
26
35
364
451
0.85 (0.57-1.27)
6.4
Pisa cohort,8 2007
39
228
238
2647
2.13 (1.52-3.00)
7.6
Whickham Survey,17,18 1996, 2010
49
124
681
2282
0.98 (0.73-1.31)
8.7
Busselton Health Study,9 2005
36
89
479
1895
1.44 (1.02-2.03)
7.6
Nagasaki Adult Health Study,34 2004
94
420
409
2171
1.04 (0.83-1.31)
10.8
Brazilian Thyroid Study,35 2010
13
101
55
876
1.96 (1.07-3.61)
3.6
915
3450
8749
51 837
1.09 (0.96-1.24)
100.0
Total (I 2 = 66%)
0.2
0.5
1
HR (95% CI)
a The sizes of the data markers are proportional to the inverse variance of the hazard ratios (HRs). CI indicates confidence interval; HUNT, Nord-Trøndelag Health
Study; HR, hazard ratio.
b Forty-six participants from the Whickham survey and 3 participants from the Busselton Health Study were not included because follow-up data were only available for death.
c Nine participants were excluded from the analysis because of missing cause of death. The Brazilian Thyroid Study was not included in this analysis because of unreliable es-
timates based on the small number of CHD deaths (n=10).
©2010 American Medical Association. All rights reserved.
(Reprinted) JAMA, September 22/29, 2010—Vol 304, No. 12
Downloaded from jama.ama-assn.org by guest on 14 December 2010
1369
SUBCLINICAL HYPOTHYROIDISM AND RISK OF CORONARY HEART DISEASE, MORTALITY
Our individual participant data analysis found that the CHD outcomes in
adults with subclinical hypothyroidism
did not differ significantly across age
groups. For the specific age group of 80
years or older, there was no significant
increased risk of total mortality, CHD
mortality, or CHD events in contrast to
a single previous study that found reduced mortality associated with increas-
10 mIU/L or greater. However, the HR
for CHD events increased to 1.28 (95%
CI, 0.68-2.39) with extended follow-up
to 8 years in the present data. In overall
pooled data, we found statistical heterogeneity among individual study findings for CHD events (I2 =59%), but not
for CHD death. Part of the heterogeneity
might be related to different CHD risks
across age, race, and TSH subgroups.
ing TSH concentrations.27,37 Previous
study-level meta-analyses have found increased risks of CHD events and cardiovascular mortality associated with subclinical hypothyroidism, particularly in
studies with a mean age of younger than
65 years,10,11 but this was not confirmed
by our individual participant data analysis. We found some evidence for increased risks of CHD events and mor-
Figure 2. Hazard Ratios (HRs) for Coronary Heart Disease (CHD) Events, CHD Mortality, and Total Mortality According to Elevated
Thyroid-Stimulating Hormone (TSH) Categories and Subclinical Hypothyroidism Stratified by Age vs Euthyroidism a
CHD Events by
TSH Level, mIU/Lb
0.5-4.49
No. of Events
4040
No. of Participants
23 957
HR Ratio
(95% CI)
1 [Reference]
4.5-6.9
264
1344
1.00 (0.86-1.18)
7.0-9.9
96
441
1.17 (0.96-1.43)
10-19.9
70
235
1.89 (1.28-2.80)
Decreased
Risk
Increased
Risk
P<.001 for trend
CHD Mortality by
TSH Level, mIU/Lc
0.5-4.49
1958
50 953
4.5-6.9
132
2363
1.09 (0.91-1.30)
7.0-9.9
50
652
1.42 (1.03-1.95)
10-19.9
28
333
1 [Reference]
1.58 (1.10-2.27)
P = .005 for trend
Total Mortality by
TSH Level, mIU/Ld
0.5-4.49
8749
51 837
4.5-6.9
640
2431
1.06 (0.96-1.17)
1 [Reference]
7.0-9.9
170
672
1.02 (0.84-1.24)
10-19.9
105
347
1.22 (0.80-1.87)
P = .39 for trend
0.2
0.5
1
2
5
HR (95% CI)
Subclinical
Hypothyroidism
CHD Events,
by Age, yb
18-49
No. of
Events
12
No. of
Participants
221
Euthyroidism
No. of
Events
272
No. of
Participants
5405
HR Ratio
(95% CI)
1.46 (0.82-2.62)
50-64
54
517
997
7876
1.13 (0.73-1.77)
65-79
322
1158
2511
9668
1.20 (0.95-1.51)
42
124
260
1008
1.30 (0.93-1.82)
≥80
CHD Mortality,
by Age, y c
18-49
2
444
54
13 560
2.13 (0.74-6.14)
16
1072
316
18 513
1.30 (0.81-2.08)
65-79
163
1608
1288
16 567
1.32 (1.08-1.62)
29
224
300
2313
1.01 (0.62-1.63)
Total Mortality,
by Age, y d
18-49
Increased
Risk
P = .78 for trend
50-64
≥80
Decreased
Risk
P = .22 for trend
14
465
340
13 832
50-64
108
1121
1492
18 875
1.17 (0.90-1.51)
65-79
623
1636
5316
16 785
1.17 (0.99-1.39)
≥80
170
228
1601
2345
0.96 (0.81-1.12)
1.31 (0.76-2.26)
P = .29 for trend
0.2
0.5
1
2
5
HR (95% CI)
a The sizes of the filled square data markers are proportional to the inverse variance of the HRs. The unfilled squares indicate the reference categories. For the analyses stratified
by age, the HRs for CHD events, CHD mortality, and total mortality were adjusted for sex and age as a continuous variable to avoid residual confounding within age strata.
CI indicates confidence interval.
b Data were available from 7 studies.
c Data were available from 10 studies. The Brazilian Thyroid Study was not included because of unreliable estimates based on the small number of CHD deaths (n=10). Nine
participants were excluded from the analysis because of missing cause of death.
d Data were available from 11 studies.
1370 JAMA, September 22/29, 2010—Vol 304, No. 12 (Reprinted)
©2010 American Medical Association. All rights reserved.
Downloaded from jama.ama-assn.org by guest on 14 December 2010
SUBCLINICAL HYPOTHYROIDISM AND RISK OF CORONARY HEART DISEASE, MORTALITY
tality in younger adults with subclinical
hypothyroidism, but there also were large
95% CIs without significant trend across
age groups (Figure 2). Moreover, the
effect of increasing TSH level on CHD
events did not significantly differ according to age. In contrast to a previous study
The increased risk of CHD events associated with higher TSH levels in our study
might be related to the known effects of
thyroid hormone on the heart and
metabolism, consistent with the concept that subclinical hypothyroidism is
a milder form of overt hypothyroid-
suggesting that adults with subclinical
hypothyroidism and preexisting cardiovascular disease might be at particularly high cardiovascular risk,8 we found
no significant effect of baseline preexisting cardiovascular disease on outcomes.
Table 2. Stratified Analyses for the Associations Between Subclinical Hypothyroidism and Risk of Coronary Heart Disease (CHD) Events, CHD
Mortality, and Total Mortality
CHD Events a
CHD Mortality b
HR (95% CI)
No. of
Events/
Total
Participants
Adjusted
for Age
and Sex
Multivariate
Model c
4470/25 977
1.18
(0.99-1.42)
Men d
2642/12 531
Women d
1828/13 446
Total population
P for interaction
Total Mortality
HR (95% CI)
No. of
Events/
Total
Participants
Adjusted
for Age
and Sex
Multivariate
Model c
1.18
(0.99-1.40)
2168/54 301
1.14
(0.99-1.32)
1.06
(0.90-1.25)
1.06
(0.91-1.25)
1246/21 889
1.21
(0.99-1.48)
1.23
(0.99-1.52)
922/32 412
.32
.27
HR (95% CI)
No. of
Events/
Total
Participants
Adjusted
for Age
and Sex
Multivariate
Model c
1.15
(0.99-1.34)
9664/55 287
1.09
(0.96-1.24)
1.13
(0.98-1.29)
1.14
(0.90-1.43)
1.12
(0.90-1.39)
4851/22 352
1.13
(0.93-1.36)
1.14
(0.93-1.39)
1.21
(0.99-1.47)
1.24
(1.01-1.53)
4813/32 935
1.06
(0.95-1.19)
1.09
(0.98-1.21)
.71
.51
.58
.70
Age, y e
18-49
284/5626
1.46
(0.82-2.62)
1.55
(0.87-2.78)
56/14 004
2.13
(0.74-6.14)
2.49
(0.87-7.19)
354/14 297
1.31
(0.76-2.26)
1.44
(0.84-2.48)
50-64
1051/8393
1.13
(0.73-1.77)
1.11
(0.75-1.66)
332/19 585
1.30
(0.81-2.08)
1.32
(0.79-2.18)
1600/19 996
1.17
(0.90-1.51)
1.22
(0.91-1.65)
65-79
2833/10 826
1.20
(0.95-1.51)
1.21
(0.96-1.52)
1451/18 175
1.32
(1.08-1.62)
1.33
(1.07-1.65)
5939/18 421
1.17
(0.99-1.39)
1.22
(1.03-1.45)
302/1132
1.30
(0.93-1.82)
1.24
(0.89-1.73)
329/2537
1.01
(0.62-1.63)
0.98
(0.60-1.60)
1771/2573
0.96
(0.81-1.12)
0.94
(0.80-1.11)
.78
.58
.22
.12
.29
.15
4193/24 746
1.20
(1.02-1.42)
1.20
(1.02-1.40)
1905/49 381
1.18
(1.01-1.38)
1.19
(1.02-1.39)
8142/49 390
1.10
(0.94-1.28)
1.11
(0.95-1.29)
Black
277/1231
0.75
(0.48-1.19)
0.73
(0.46-1.17)
108/1231
0.67
(0.31-1.44)
0.59
(0.25-1.37)
484/1231
0.94
(0.69-1.29)
0.96
(0.70-1.32)
Asian
NA
NA
NA
31/2591
0.67
(0.23-1.91)
0.67
(0.23-1.95)
571/3568
1.34
(0.73-2.46)
1.39
(0.78-2.46)
.05
.05
.23
.18
.52
.51
ⱖ80
P for trend
Race f
White
P for interaction
TSH, mIU/L
0.5-4.49
4040/23 957
1 [Reference]
1 [Reference]
1958/50 953
1 [Reference]
1 [Reference]
8749/51 837
1 [Reference]
1 [Reference]
4.5-6.9
264/1344
1.00
(0.86-1.18)
1.01
(0.86-1.18)
132/2363
1.09
(0.91-1.30)
1.06
(0.88-1.28)
640/2431
1.06
(0.96-1.17)
1.07
(0.96-1.20)
7.0-9.9
96/441
1.17
(0.96-1.43)
1.22
(0.99-1.49)
50/652
1.42
(1.03-1.95)
1.53
(1.13-2.07)
170/672
1.02
(0.84-1.24)
1.11
(0.92-1.33)
10.0-19.9
70/235
1.89
(1.28-2.80)
1.86
(1.22-2.82)
28/333
1.58
(1.10-2.27)
1.54
(1.07-2.23)
105/347
1.22
(0.80-1.87)
1.24
(0.82-1.87)
⬍.001
.002
.005
.005
.39
.29
1282/4263
1.17
(0.94-1.47)
1.09
(0.90-1.31)
590/4390
1.30
(0.98-1.72)
1.28
(0.99-1.66)
1649/4523
1.08
(0.87-1.34)
1.05
(0.86-1.29)
3142/21 391
1.16
(0.95-1.40)
1.18
(0.97-1.43)
1450/48 776
1.08
(0.89-1.30)
1.10
(0.91-1.33)
7532/49 629
1.10
(0.95-1.28)
1.13
(0.97-1.31)
.96
.57
.29
.35
.89
.57
P for trend
Cardiovascular disease g
Yes
No
P for interaction
Abbreviations: CI, confidence interval; HR, hazard ratio; NA, data not applicable; TSH, thyroid-stimulating hormone.
a Data were available from 7 studies. Forty-six participants from the Whickham survey and 3 participants from the Busselton Health Study were not included in the analysis of CHD events
because follow-up data were only available for death.
b Nine participants were excluded from this analysis because of missing cause of death. The Brazilian Thyroid Study was not included in this analysis because of unreliable estimates due
to the low number of CHD deaths (n=10).
c Adjusted for sex, age, systolic blood pressure, current and former smoking, total cholesterol, and prevalent diabetes at baseline. The Birmingham Study was not included in this analysis
because of lack of data on cardiovascular risk factors.
d These HRs were not adjusted for sex.
e These HRs were adjusted for sex and age as a continuous variable to avoid residual confounding within age strata.
f Data were not available for the Birmingham study (n=1098).
g Data were not available for the Birmingham study (n=1098). Thirty-seven participants with missing information on baseline cardiovascular disease from other studies were excluded from
this analysis. For analysis of CHD events, 286 participants without preexisting cardiovascular disease from the Leiden 85-plus Study were further excluded because of no CHD event.
©2010 American Medical Association. All rights reserved.
(Reprinted) JAMA, September 22/29, 2010—Vol 304, No. 12
Downloaded from jama.ama-assn.org by guest on 14 December 2010
1371
SUBCLINICAL HYPOTHYROIDISM AND RISK OF CORONARY HEART DISEASE, MORTALITY
ism.38,39 Increased systemic vascular resistance, arterial stiffness, altered endothelial function, increased atherosclerosis,
and altered coagulability have been
reported to be associated with subclinical hypothyroidism and may accelerate
development of CHD.4,39,40 The fact that
adjustment for traditional cardiovascu-
lar risk factors did not alter risks could
favor this hypothesis. Other potential
mechanisms include elevated cholesterol level,4,39 although adjustment for
Table 3. Sensitivity Analysis of the Effect of Subclinical Hypothyroidism on the Risk of Coronary Heart Disease (CHD) Events and CHD Mortality a
CHD Events by Thyroid-Stimulating
Hormone Level, mIU/L b
CHD Mortality by Thyroid-Stimulating
Hormone Level, mIU/L
Subclinical Hypothyroidism
Subclinical Hypothyroidism
No. of
No. of
4.5-19.9
10-19.9
4.5-19.9
10-19.9
Events/
Events/
Participants
Participants
No. of
No. of
No. of
No. of
With
With
Events/
HR
Events/
HR
HR
Events/
HR
Euthyroidism,
Euthyroidism, Events/
Participants (95% CI) Participants (95% CI)
0.5-4.49
0.5-4.49 Participants (95% CI) Participants (95% CI)
Random-effects model
4040/23 957
430/2020
1.18
70/235
1.89
1958/50 953 210/3348
1.14
28/333
1.58
(0.99-1.42)
(1.28-2.80)
(0.99-1.32)
(1.10-2.27)
Fixed-effects model
4040/23 957
430/2020
1.10
70/235
1.81
1958/50 953 210/3348
1.14
28/333
1.58
(0.99-1.22)
(1.43-2.30)
(0.99-1.32)
(1.10-2.27)
Excluding those with subclinical hypothyroidism
With thyroid
medication use c
At baseline
3972/23 682
412/1937
1.16
60/204
1.77
1924/50 653 204/3253
1.14
24/300
1.46
(0.97-1.38)
(1.13-2.76)
(0.99-1.32)
(0.99-2.17)
At baseline
2354/11 635
246/998
1.17
29/73
2.17
1114/14 829 130/1466
1.25
15/90
1.85
and
(0.91-1.50)
(1.19-3.93)
(1.04-1.51)
(1.13-3.05)
during
follow-up
With missing free
4040/23 957
423/1995
1.19
70/232
1.85
1958/50 953 204/3303
1.15
28/330
1.55
thyroxine (T4) d
(0.99-1.42)
(1.22-2.80)
(0.99-1.33)
(1.07-2.25)
Excluding soft CHD
outcomes e
3393/23 957
334/2020
Studies with formal
adjudication
procedures6-8,27, f
1672/7932
269/1090
1.23
(1.04-1.46)
1.08
(0.85-1.37)
Outcomes
53/235
1.81
(1.10-2.98)
41/113
2.05
(1.14-3.68)
NA
NA
654/7929
111/1089
1.13
(0.83-1.55)
16/112
1.77
(1.08-2.89)
Adjustments g
Cardiovascular risk factors
Plus lipid-lowering
2465/12 060
and
antihypertensive
medications
Plus BMI
4040/23 957
Study of cardiac
patients8
3892/21 310
327/1300
1.23
(0.97-1.57)
55/155
1.90
(1.09-3.34)
1396/35 879
164/2116
1.14
(0.96-1.35)
24/236
1.57
(1.04-2.37)
430/2020
1.16
(0.98-1.37)
70/235
1.86
(1.22-2.85)
1845/49 947
199/3256
1.13
(0.97-1.32)
28/316
1.45
(0.99-2.13)
410/1792
1.13
(0.95-1.34)
NA i
1866/48 306
196/3120
26/310
1931/48 782
206/2928
1583/27 177
186/2534
1.10
(0.95-1.28)
1.15
(1.00-1.34)
1.15
(0.99-1.34)
1.53
(1.05-2.23) h
1.57
(1.09-2.26)
1.61
(1.12-2.33)
Atomic bomb survivors
in Nagasaki, Japan34
HUNT Study33, j
Rotterdam Study,5, k
NA i
4050/24 807
434/2127
Studies Excluded
64/212
1.66
(1.19-2.31) h
NA i
NA i
Additional Study Considered
1.20
NA l
(1.00-1.44)
NA l
28/318
28/268
NA l
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); CI, confidence interval; HR, hazard ratio; NA, data not applicable.
a The HRs were adjusted for age and sex using a random-effects model.
b Data were available from 7 studies.
c The numbers of participants with thyroid medication use appear in columns 7 and 8 of Table 1. The HUNT Study and the EPIC-Norfolk Study were not included in this analysis because
of lack of this information during follow-up.
d The numbers of participants appear in the eTable at http://www.jama.com.
e Defined as hospitalization for angina or revascularization (coronary angioplasty or surgery) and participants with these outcomes were excluded from this analysis, which was possible
for participants from 4 studies (eTable). In contrast, hard events were defined as nonfatal myocardial infarction or CHD death, as defined in the current Framingham risk score.20
f Defined as having clear criteria for the outcomes that were reviewed by experts for each potential case (eg, specific electrocardiogram or cardiac enzymes modifications for CHD). For
this analysis, CHD adjudication based only on death certificates was not considered as a formal adjudication procedure.
g The Birmingham Study was excluded from these analyses because of lack of data on cardiovascular risk factors. Data on lipid-lowering and antihypertensive medications were not
available for the EPIC-Norfolk and Nagasaki Adult Health studies.
h With further adjustment for cardiovascular risk factors after excluding the Pisa cohort, the HRs for TSH level of 10-19.9 mIU/L were 1.63 (95% CI, 1.13-2.34) for CHD events, 1.52 (95%
CI, 1.04-2.23) for CHD mortality, and 1.05 (95% CI, 0.79-1.40) for total mortality (vs an HR of 1.06 [95% CI, 0.83-1.35] in age- and sex-adjusted analyses excluding the Pisa cohort).
i No data on CHD events were available.
j Had the lowest rate of subclinical hypothyroidism (3.3%, Table 1).
k This study had 14 CHD events5,10 but did not accept invitation to share individual participant data. Summary estimates of this study, adjusted for age, BMI, total cholesterol, high-density
lipoprotein cholesterol, blood pressure, and smoking were used in the random-effect models as a sensitivity analysis.24
l The TSH subgroups were not reported in the study.
1372 JAMA, September 22/29, 2010—Vol 304, No. 12 (Reprinted)
©2010 American Medical Association. All rights reserved.
Downloaded from jama.ama-assn.org by guest on 14 December 2010
SUBCLINICAL HYPOTHYROIDISM AND RISK OF CORONARY HEART DISEASE, MORTALITY
cholesterol level did not remove the associations in our data. Adults with higher
TSH concentrations also are more likely
to develop overt hypothyroidism,41 and
it is possible that this progression explains
the association with subclinical hypothyroidism. Alternative explanations for
the observed results are bias in the selection of included studies, bias and quality problems in the original studies,
publication bias, and unmeasured confounders.42 Sensitivity analyses pooling
higher-quality studies yielded similar
results. Whereas one randomized controlled trial has shown benefits with thyroxine treatment of subclinical hypothyroidism on intima-media thickness40 and
another has shown benefits with thyroxine treatment of subclinical hypothyroidism on brachial artery endothelial
function,43 the potential causal relationship can only be proven by randomized
controlled trials of thyroxine replacement and clinical outcomes.36
Among the strengths of our study, an
individual participant data analysis is the
preferred way to perform time-to-event
analyses to avoid biases associated with
the use of aggregate data in metaregression for subgroup analysis and to
allow standardization of definitions of
predictors, outcomes, and adjustment for
potential confounders.14,22 We included
all available international and published data on these associations. Among
the limitations of our study, the individual participant data analysis included
predominantly white populations, except
for 2 studies conducted in Japan34 and
Brazil.35 Results for subgroups at risk of
CHD mortality generally had wider 95%
CIs than those for CHD events, reflecting less statistical power. However, post
hoc calculations showed 80% power to
detect meaningful differences between
overall subclinical hypothyroidism and
euthyroidism groups for each outcome.
Specifically, our study had adequate
power to detect an HR of 1.18 or higher
for CHD events, an HR of 1.30 or higher
for CHD mortality, and an HR of 1.13 or
higher for total mortality. Even with this
very large amount of individual participant data, our power for subgroup analyses was limited among those with TSH
levels of 10 mIU/L or greater or adults
younger than 50 years because of the limited number of CHD events and deaths.
Thyroid function testing was performed only at baseline, and we have no
data on how many participants progressed from euthyroidism to subclinical hypothyroidism, from subclinical to
overt hypothyroidism, or who normalized their TSH level over time, which is
a limitation of all published large
cohorts.6,7,33 In addition, free triiodothyronine (T3) was not available in most
cohorts, and thus could not be included
in thyroid status classification. Commencement of thyroid medication during follow-up by up to 12.6% of participants might have attenuated any true
effects of subclinical hypothyroidism, as
illustrated by the sensitivity analysis
excluding such participants.
In summary, combining all available
data from large prospective cohorts
among 55 287 individual participants
suggests that subclinical hypothyroidism is associated with an increased risk
of CHD in those with higher TSH levels.
The risk of both CHD mortality and CHD
events, but not of total mortality,
increases with higher concentrations of
TSH and is significantly elevated in adults
with TSH levels of 10 mIU/L or greater.
Conversely, minimal TSH elevations are
not associated with an increased risk of
CHD events and CHD mortality. Our
finding of no increased risk of CHD
among the high proportions of adults
with minimal TSH elevations is also
important because many patients with
minimal TSH elevations are currently
treated in clinical practice.44 Our results
might help refine a TSH threshold at
which larger clinical benefits of thyroxine replacement would be expected.4,45
Our study cannot address whether these
risks are attenuated or abolished by thyroxine replacement. Given the high
prevalence of subclinical hypothyroidism, 2 , 1 9 this question needs to be
addressed in an appropriately powered
randomized controlled trial.
Author Affiliations: Department of Ambulatory Care and
Community Medicine, University of Lausanne, Lausanne, Switzerland (Drs Rodondi, Collet, and Cornuz);
Departments of Public Health and Primary Care (Ms den
Elzen and Dr Gussekloo) and Gerontology and Geri-
©2010 American Medical Association. All rights reserved.
atrics (Dr Westendorp), Leiden University Medical Center, Leiden, the Netherlands; Departments of Medicine, Epidemiology, and Biostatistics (Drs Bauer and
Vittinghoff ) and Medicine (Dr Bauer), University of California, San Francisco; Division of Endocrinology, Diabetes, and Metabolism, Department of Medicine, School
of Medicine, University of Pennsylvania, Philadelphia (Dr
Cappola); Department of Endocrinology, Gateshead
Health Foundation NHS Trust, Gateshead, England (Dr
Razvi); Department of Endocrinology and Diabetes, Sir
Charles Gairdner Hospital, Nedlands, Western Australia (Dr Walsh); Schools of Medicine and Pharmacology
(Dr Walsh) and Population Health (Dr Bremner), University of Western Australia, Crawley; Department of
Public Health, Norwegian University of Science and Technology, Trondheim, Norway (Dr Åsvold); National Council Research Institute of Clinical Physiology, Pisa, Italy
(Dr Iervasi); Department of Clinical Studies, Radiation
Effects Research Foundation, Nagasaki, Japan (Dr Imaizumi); Division of Epidemiology and Biostatistics, European Institute of Oncology, Milan, Italy (Mr Maisonneuve); Division of Endocrinology, Department of
Medicine, Federal University of Sao Paulo, Brazil (Dr
Sgarbi); Division of Endocrinology, Faculdade de Medicina de Marı́lia, Marı́lia, Brazil (Dr Sgarbi); Department of Public Health and Primary Care, University of
Cambridge, Cambridge, England (Dr Khaw); Department of Endocrinology, Royal Free Hospital, London,
England (Dr Vanderpump); Department of Epidemiology, University of Pittsburgh, Pittsburgh, Pennsylvania
(Dr Newman); School of Clinical and Experimental Medicine, College of Medical and Dental Sciences, University of Birmingham, Birmingham, England (Dr Franklyn); and the Netherlands Consortium for Healthy
Ageing, Leiden (Dr Westendorp).
Author Contributions: Dr Rodondi had full access to
all of the data in the study and takes responsibility for
the integrity of the data and the accuracy of the data
analysis.
Study concept and design: Rodondi, Bauer, Cornuz,
Westendorp, Gussekloo.
Acquisition of data: Rodondi, Bauer, Walsh, Åsvold,
Iervasi, Imaizumi, Sgarbi, Khaw, Vanderpump,
Newman, Franklyn, Westendorp, Gussekloo.
Analysis and interpretation of data: Rodondi, den Elzen,
Bauer, Cappola, Razvi, Åsvold, Iervasi, Imaizumi, Collet,
Bremner, Maisonneuve, Sgarbi, Cornuz, Franklyn,
Westendorp, Vittinghoff, Gussekloo.
Drafting of the manuscript: Rodondi.
Critical revision of the manuscript for important intellectual content: den Elzen, Bauer, Cappola, Razvi,
Walsh, Åsvold, Iervasi, Imaizumi, Collet, Bremner,
Maisonneuve, Sgarbi, Khaw, Vanderpump, Newman,
Cornuz, Franklyn, Westendorp, Vittinghoff, Gussekloo.
Statisticalanalysis:Rodondi,denElzen,Bauer,Vittinghoff.
Obtained funding: Walsh, Iervasi, Sgarbi, Khaw,
Vanderpump, Newman, Franklyn, Westendorp,
Gussekloo.
Administrative, technical, or material support:
Rodondi, Collet, Khaw, Newman, Gussekloo.
Study supervision: Rodondi, Westendorp, Gussekloo.
Financial Disclosures: None reported.
Funding/Support: The Cardiovascular Health Study and
the research reported in this article were supported by
contract numbers N01-HC-80007, N01-HC-85079
through N01-HC-85086, N01-HC-35129, N01 HC15103, N01 HC-55222, N01-HC-75150, N01-HC45133, grant number U01 HL080295 from the National Heart, Lung, and Blood Institute, with additional
funding from the National Institute of Neurological Disorders and Stroke. Additional support was provided
through grants R01 AG-15928, R01 AG-20098, AG027058, and AG-032317 from the National Institute
on Aging, grant R01 HL-075366 from the National
Heart, Lung, and Blood Institute, and grant P30-AG024827 from the University of Pittsburgh Claude. D.
Pepper Older Americans Independence Center. A full
list of principal investigators and institutions of the Car-
(Reprinted) JAMA, September 22/29, 2010—Vol 304, No. 12
Downloaded from jama.ama-assn.org by guest on 14 December 2010
1373
SUBCLINICAL HYPOTHYROIDISM AND RISK OF CORONARY HEART DISEASE, MORTALITY
diovascular Health Study can be found at http://www
.chs-nhlbi.org/pi.htm. The thyroid measurements in the
Cardiovascular Health Study were supported by an
American Heart Association Grant-in-Aid (to Linda
Fried). The Health, Aging, and Body Composition Study
is supported by National Institute on Aging contract
numbers N01-AG-6-2101, N01-AG-6-2103, and N01AG-6-2106, and in part by the Intramural Research Program of the National Institutes of Health. The National Institute on Aging funded the Health Aging, and
Body Composition study. The Leiden 85-plus Study was
partly funded by the Dutch Ministry of Health, Welfare, and Sports. The Whickham Survey was supported by the UK Department of Health. The HUNT
Study was a collaborative effort of the Faculty of Medicine, Norwegian University of Science and Technology, the Norwegian Institute of Public Health, and the
Nord-Trøndelag County Council. The thyroid testing
in the HUNT Study was financially supported by Wallac Oy (Turku, Finland). The Nagasaki Adult Health
Study was supported by the Radiation Effects Research Foundation, Hiroshima and Nagasaki, Japan, a
private, nonprofit foundation funded by the Japanese
Ministry of Health, Labor and Welfare and the US Department of Energy, the latter in part through the National Academy of Sciences. This publication was supported by research protocol A-10-08 from the Radiation
Effects Research Foundation. The EPIC-Norfolk study
was supported by research grants from the UK Medical Research Council and the UK Cancer Research. The
Brazilian Thyroid Study was supported by an unrestricted grant from the Sao Paulo State Research Foundation (Fundacão de Amparo a⬘ Pesquisa do Estado de
Sao Paulo grant 6/59737-9 to Rui Maciel). Dr Newman is supported by grant AG-023629 from the the
National Institute on Aging. Dr Westendorp is supported by grant NGI/NWO 911-03-016 from the Netherlands Organization for Scientific Research.
Role of the Sponsor: The majority of the sponsors had
no role in the design and conduct of the study; collection, management, analysis, and interpretation of
the data; or preparation, review, or approval of the
manuscript. The National Institute on Aging funded
the Health, Aging, and Body Composition study and
reviewed the manuscript and approved its publication. The Radiation Effects Research Foundation funded
the Nagasaki Adult Health Study and reviewed the
manuscript and approved its publication.
Statistical Evaluation: Dr Vittinghoff, professor of biostatistics, in the Department of Epidemiology and Biostatistics, University of California, San Francisco, reviewed the statistical analyses of the article.
Participating Studies of the Thyroid Studies Collaboration: United States: Cardiovascular Health Study;
Health, Aging, and Body Composition Study. The Netherlands: the Leiden 85-plus Study. Australia: Busselton Health Study. United Kingdom: Whickham Survey; Birmingham Study; EPIC-Norfolk Study. Italy: Pisa
Cohort. Japan: Nagasaki Adult Health Study. Brazil:
Brazilian Thyroid Study. Norway: Nord-Trøndelag
Health Study (HUNT Study).
Online-Only Material: eMethods, eTable, and eFigure are available at http://www.jama.com.
Additional Contributions: We thank Sabrina Molinaro
(Clinical Physiology Institute, Pisa, Italy) for technical help
about data from the Pisa Cohort and from Rui Maciel
(Escola Paulista de Medicina, Federal University of Sao
Paulo, Brazil) for technical help about data from the Brazilian Thyroid Study. The persons listed in this section
did not receive financial compensation.
REFERENCES
1. Helfand M; US Preventive Services Task Force.
Screening for subclinical thyroid dysfunction in nonpregnant adults. Ann Intern Med. 2004;140(2):
128-141.
2. Surks MI, Ortiz E, Daniels GH, et al. Subclinical thyroid disease. JAMA. 2004;291(2):228-238.
3. Gharib H, Tuttle RM, Baskin HJ, et al. Subclinical
thyroid dysfunction. J Clin Endocrinol Metab. 2005;
90(1):581-585.
4. Biondi B, Cooper DS. The clinical significance of subclinical thyroid dysfunction. Endocr Rev. 2008;
29(1):76-131.
5. Hak AE, Pols HA, Visser TJ, et al. Subclinical hypothyroidism is an independent risk factor for atherosclerosis and myocardial infarction in elderly women.
Ann Intern Med. 2000;132(4):270-278.
6. Cappola AR, Fried LP, Arnold AM, et al. Thyroid
status, cardiovascular risk, and mortality in older adults.
JAMA. 2006;295(9):1033-1041.
7. Rodondi N, Newman AB, Vittinghoff E, et al. Subclinical hypothyroidism and the risk of heart failure,
other cardiovascular events, and death. Arch Intern
Med. 2005;165(21):2460-2466.
8. Iervasi G, Molinaro S, Landi P, et al. Association
between increased mortality and mild thyroid dysfunction in cardiac patients. Arch Intern Med. 2007;
167(14):1526-1532.
9. Walsh JP, Bremner AP, Bulsara MK, et al. Subclinical thyroid dysfunction as a risk factor for cardiovascular disease. Arch Intern Med. 2005;165(21):
2467-2472.
10. Ochs N, Auer R, Bauer DC, et al. Meta-analysis:
subclinical thyroid dysfunction and the risk for coronary heart disease and mortality. Ann Intern Med.
2008;148(11):832-845.
11. Razvi S, Shakoor A, Vanderpump M, et al. The
influence of age on the relationship between subclinical hypothyroidism and ischemic heart disease. J Clin
Endocrinol Metab. 2008;93(8):2998-3007.
12. Völzke H, Schwahn C, Wallaschofski H, Dörr M.
Review: the association of thyroid dysfunction with
all-cause and circulatory mortality. J Clin Endocrinol
Metab. 2007;92(7):2421-2429.
13. Egger M, Davey Smith G, Altman D. Systematic
Reviews in Health Care: Meta-analysis in Context.
London, England: BMJ Publishing Group; 2001.
14. Simmonds MC, Higgins JP, Stewart LA, et al. Metaanalysis of individual patient data from randomized
trials. Clin Trials. 2005;2(3):209-217.
15. Rodondi N, Bauer DC, Cappola AR, et al. Subclinical thyroid dysfunction, cardiac function, and the
risk of heart failure. J Am Coll Cardiol. 2008;52
(14):1152-1159.
16. Nicoloff JT, Spencer CA. Clinical review 12: the
use and misuse of the sensitive thyrotropin assays.
J Clin Endocrinol Metab. 1990;71(3):553-558.
17. Vanderpump MP, Tunbridge WM, French JM,
et al. The development of ischemic heart disease in
relation to autoimmune thyroid disease in a 20-year
follow-up study of an English community. Thyroid.
1996;6(3):155-160.
18. Razvi S, Weaver JU, Vanderpump MP, Pearce SH.
The incidence of ischemic heart disease and mortality
in people with subclinical hypothyroidism: reanalysis
of the Whickham Survey cohort. J Clin Endocrinol
Metab. 2010;95(4):1734-1740.
19. Hollowell JG, Staehling NW, Flanders WD, et al.
Serum TSH, T(4), and thyroid antibodies in the United
States population (1988 to 1994). J Clin Endocrinol
Metab. 2002;87(2):489-499.
20. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive
Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on
Detection, Evaluation, and Treatment of High Blood
Cholesterol in Adults (Adult Treatment Panel III). JAMA.
2001;285(19):2486-2497.
21. DerSimonian R, Laird N. Meta-analysis in clinical
trials. Control Clin Trials. 1986;7(3):177-188.
22. Stewart LA, Clarke MJ; Cochrane Working Group.
Practical methodology of meta-analyses (overviews)
using updated individual patient data. Stat Med. 1995;
14(19):2057-2079.
23. Fowkes FG, Murray GD, Butcher I, et al; Ankle
1374 JAMA, September 22/29, 2010—Vol 304, No. 12 (Reprinted)
Brachial Index Collaboration. Ankle brachial index combined with Framingham Risk Score to predict cardiovascular events and mortality. JAMA. 2008;300
(2):197-208.
24. Riley RD, Simmonds MC, Look MP. Evidence synthesis combining individual patient data and aggregate data. J Clin Epidemiol. 2007;60(5):431-439.
25. Higgins JPT, Thompson SG, Deeks JJ, Altman DG.
Measuring inconsistency in meta-analyses. BMJ. 2003;
327(7414):557-560.
26. Parle JV, Maisonneuve P, Sheppard MC, et al. Prediction of all-cause and cardiovascular mortality in elderly people from one low serum thyrotropin result.
Lancet. 2001;358(9285):861-865.
27. Gussekloo J, van Exel E, de Craen AJ, et al. Thyroid status, disability and cognitive function, and survival in old age. JAMA. 2004;292(21):2591-2599.
28. Heinze G, Schemper M. A solution to the problem of monotone likelihood in Cox regression.
Biometrics. 2001;57(1):114-119.
29. Sawin CT, Geller A, Wolf PA, et al. Low serum
thyrotropin concentrations as a risk factor for atrial fibrillation in older persons. N Engl J Med. 1994;
331(19):1249-1252.
30. Vittinghoff E, Glidden D, Shiboski S, McCulloch
C. Regression Methods in Biostatistics: Linear, Logistic, Survival, and Repeated Measures Models. New
York, NY: Springer-Verlag; 2005.
31. Egger M, Davey Smith G, Schneider M, Minder
C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629-634.
32. Boekholdt SM, Titan SM, Wiersinga WM, et al.
Initial thyroid status and cardiovascular risk factors. Clin
Endocrinol (Oxf ). 2010;72(3):404-410.
33. Asvold BO, Bjøro T, Nilsen TI, Gunnell D, Vatten
LJ. Thyrotropin levels and risk of fatal coronary heart
disease. Arch Intern Med. 2008;168(8):855-860.
34. Imaizumi M, Akahoshi M, Ichimaru S, et al. Risk
for ischemic heart disease and all-cause mortality in
subclinical hypothyroidism. J Clin Endocrinol Metab.
2004;89(7):3365-3370.
35. Sgarbi JA, Matsumura LK, Kasamatsu TS, et al.
Subclinical thyroid dysfunctions are independent risk
factors for mortality in a 7.5-year follow-up. Eur J
Endocrinol. 2010;162(3):569-577.
36. Ladenson PW. Cardiovascular consequences of
subclinical thyroid dysfunction. Ann Intern Med. 2008;
148(11):880-881.
37. Cooper DS. Thyroid disease in the oldest old.
JAMA. 2004;292(21):2651-2654.
38. Klein I, Ojamaa K. Thyroid hormone and the cardiovascular system. N Engl J Med. 2001;344(7):
501-509.
39. Klein I, Danzi S. Thyroid disease and the heart.
Circulation. 2007;116(15):1725-1735.
40. Monzani F, Caraccio N, Kozàkowà M, et al. Effect
of levothyroxine replacement on lipid profile
and intima-media thickness in subclinical
hypothyroidism. J Clin Endocrinol Metab. 2004;
89(5):2099-2106.
41. Vanderpump MP, Tunbridge WM, French JM,
et al. The incidence of thyroid disorders in the
community. Clin Endocrinol (Oxf ). 1995;43(1):
55-68.
42. Stroup DF, Berlin JA, Morton SC, et al. Metaanalysis of observational studies in epidemiology.
JAMA. 2000;283(15):2008-2012.
43. Razvi S, Ingoe L, Keeka G, et al. The beneficial
effect of L-thyroxine on cardiovascular risk factors, endothelial function, and quality of life in subclinical
hypothyroidism. J Clin Endocrinol Metab. 2007;
92(5):1715-1723.
44. Fatourechi V, Lankarani M, Schryver PG, et al. Factors influencing clinical decisions to initiate thyroxine
therapy for patients with mildly increased serum thyrotropin (5.1-10.0 mIU/L). Mayo Clin Proc. 2003;
78(5):554-560.
45. Cappola AR. Subclinical thyroid dysfunction and
the heart. J Clin Endocrinol Metab. 2007;92(9):
3404-3405.
©2010 American Medical Association. All rights reserved.
Downloaded from jama.ama-assn.org by guest on 14 December 2010
ANEXO 5
Trabalho No. 05
Associations between subclinical hypothyroidism and traditional and
nontraditional cardiovascular risk factors
José A. Sgarbi, Luiza K. Matsumura, Teresa S. Kasamatsu, Heloisa H. Villar,
Sandra R. Ferreira, Rui M. B. Maciel.
Manuscristo em preparo para publicação
70
ASSOCIATIONS BETWEEN SUBCLINICAL HYPOTHYROIDISM AND
TRADITIONAL AND NONTRADITIONAL CARDIOVASCULAR RISK
FACTORS
José A. Sgarbi, MD; Luiza K. Matsumura, MD, PhD; Teresa S. Kasamatsu, MS;
Villar H, MD, PhD; Sandra R. Ferreira, MD, PhD; Rui M. B. Maciel , MD, PhD
Author affiliations
1. Division of Endocrinology, Department of Medicine, Escola Paulista de
Medicina, Federal University of São Paulo, São Paulo, Brazil (Drs. Sgarbi,
Matsumura, Ferreira and Maciel, and Ms. Kasamatsu)
2. Division of Endocrinology, Department of Medicine, Marília School of
Medicine, Marília, Brazil (Dr Sgarbi, Dr. Villar)
3. Department of Nutrition, School of Public Health, University of São Paulo,
São Paulo, Brazil (Dr Ferreira)
Corresponding author
Rui M. B. Maciel, MD, PhD
Laboratory of Molecular Endocrinology, Division of Endocrinology, Department
of Medicine, Escola Paulista de Medicina, Federal University of São Paulo
Rua Pedro de Toledo 781, 12o. andar, 04029-032, São Paulo, SP, Brazil
Phone: (+55) 11-5084-5231 e-mail: [email protected]
Grant
The Japanese-Brazilian Thyroid Study was supported by an unrestricted grant
from FAPESP – Fundação de Amparo à Pesquisa do Estado de São Paulo
(06/59737-9)
Disclosure of Potential Conflict of Interest
The authors declare that there is no conflict of interest that could be perceived
as prejudicing the impartiality of the research reported.
71
Key words: subclinical hypothyroidism, cardiovascular risk factors, metabolic
syndrome
72
Abstract
Context: Subclinical hypothyroidism (SCH) has been associated with coronary
heart disease (CHD) and mortality, but the causes of CHD remain unknown.
Possible causes include cardiovascular (CV) risk factors related to metabolic
syndrome (MS), but the available data are conflicting.
Objective: To investigate the hypothesis that SCH is associated with traditional
and nontraditional CV risk factors and that SCH is associated with a cluster of
these risk factors in a population with a high prevalence of MS.
Design: Cross-sectional study
Setting: Population-based study
Participants: A Japanese-Brazilian population aged more than 30 yr, free of
thyroid disease and thyroid medication, was studied. Individuals with SCH (n =
99) were compared with those with euthyroidism (n = 913).
Measurements: Clinical signs of MS, both traditional (smoking, diabetes,
hypertension, and low-density lipoprotein cholesterol) and nontraditional (waist
circumference,
triglycerides,
high-density
lipoprotein
cholesterol,
insulin
resistance, apolipoprotein B, C-reactive protein, interleukin-6, homocysteine,
and uric acid), and a cluster of these factors.
Results: SCH was associated with higher concentrations of both triglycerides
and homocysteine in women, but the differences in these markers were not
73
significant. SCH was not associated with MS, other traditional or nontraditional
risk factors, or with a cluster of these factors in this population.
Conclusions: These findings suggest that the increased risk of CHD events
and CHD mortality associated with SCH might be associated with mechanisms
other than insulin resistance and low-grade inflammation.
74
Introduction
Subclinical hypothyroidism (SCH) has been defined as a mild thyroid
failure characterized by normal serum levels of thyroid hormones in the face of
mildly elevated serum thyroid-stimulating hormone TSH concentrations (1).
Epidemiological studies show a high prevalence of undiagnosed SCH in the
general population, particularly in women at older ages, with rates ranging from
10% to 15% (2-4). Despite this high prevalence and the more frequent
diagnosis of this condition in daily clinical practice, there is no consensus
regarding screening and indications for the treatment of SCH (5-7).
The clinical significance of SCH remains controversial, specifically
whether it is independently associated with coronary heart disease (CHD) and
mortality. In the last decade, large prospective observational studies (8-11) and
meta-analyses (12-14) have reported conflicting results. Most recently, an
individual participant data analysis (15) from 11 prospective cohort studies
found that SCH was significantly associated with an increased risk of CHD
events and CHD mortality, but the reasons for this association remain unknown
(1). Possible reasons include accelerated atherosclerosis (8), disturbance in
atherogenic lipid metabolism (16-18), endothelium dysfunction (19), and altered
coagulation parameters (20).
More recently, studies have speculated a possible relationship between
SCH and metabolic syndrome (MS) (21-24) or components of MS (25,26), but
the data are inconsistent. Considering that MS represents a combination of
multiple risk determinants, including traditional and nontraditional cardiovascular
(CV) risk factors (27,28), that the metabolic abnormalities and increased CV
described in MS are similar to those seen in SCH (21,23), and that the few
75
available data on this matter are conflicting, it is of interest to investigate
potential relationships between SCH and MS or its components in a population
with a high prevalence of MS.
We therefore investigated the possible associations between SCH and
MS, traditional and nontraditional CV risk factors, and a cluster of these factors
in a population of Japanese ancestry noted for its high prevalence of MS
(29,30).
Methods
Study population
The participants of this study were part of the Japanese-Brazilian
Diabetes Study designed to investigate the prevalence of diabetes and
associated diseases in an urban community of Japanese ancestry living in
Bauru, a developed city of São Paulo State in Brazil. Details on the selection
and recruitment of the sample population have been previously published (29).
We also previously reported the prevalence of thyroid dysfunction and the
association of subclinical thyroid disorders with mortality in this population (11).
Among the 1330 participants ≥ 30 years of age who agreed to participate, we
excluded those who self-reported thyroid disease or were taking thyroid
medications; those using amiodarone, lithium, or corticosteroids; and those
without a complete cardiovascular evaluation. Participants with other thyroid
dysfunctions, except SCH, and those with atypical thyroid function tests not
matching any category defined for the present study were also excluded from
our study. In total, 1012 individuals were analyzed. There were no differences
in the demographic or clinical characteristics between included and excluded
76
(n=318) individuals. All participants gave written informed consent, and the
study design was approved by the Ethics Committee of Escola Paulista de
Medicina, Federal University of São Paulo.
Study procedures
Sociodemographic, cultural, lifestyle, and health data were obtained by
standardized questionnaires and trained interviewers. A specific thyroid
questionnaire that included family and personal histories of thyroid disease was
applied by experts in thyroid diseases.
Body weight and height were measured while individuals were wearing
light clothing without shoes. Waist circumference was measured at the level of
the umbilicus while standing and during slight expiration. Blood pressure was
taken three times by an automatic device (Omron model HEM-712C, Omron
Health Care, USA). The mean of the last two measurements was used to
express systolic and diastolic blood pressure values. A standard 12-lead
electrocardiogram (ECG) was obtained in the resting state by the standard
procedure and analyzed by two cardiologists.
Fasting blood samples were taken, and a 75 g oral glucose tolerance
test was performed. Samples were processed for immediate analysis in the
local laboratory or stored at −80 C. Plasma glucose was measured by the
glucose oxidase method, while the total cholesterol, high-density lipoprotein
cholesterol (HDL-c), and triglycerides were enzymatically evaluated by an
automatic analyzer. Low-density lipoprotein cholesterol (LDL-c) levels were
calculated according the Friedewald equation (31). Apolipoprotein B (apo B)
levels were determined by an immunoturbidimetric assay (normal range, 66–
77
133 mg/dL for men and 60–117 mg/dL for women). C-reactive protein (CRP;
normal range, 0.05–0.11 mg/dL) and interleukin-6 (IL-6; normal range, up to 9.7
pg/mL) levels were determined by a chemiluminescent immunometric assay.
Serum homocysteine levels were determined by high-performance liquid
chromatography, as described by Pfeiffer et al. (normal range, 5–15 µmol/L)
(32). Albumin and creatinine levels were quantified from an early-morning urine
specimen, and the albumin/creatinine ratio was calculated using the DCA2000®
microalbumin/creatinine
assay
system
(Bayer
Diagnostica,
Leverkusen,
Germany), which detects albumin by immunoturbidimetric direct antibodyantigen aggregation and measures creatinine colorimetrically using the
Benedict-Behre reaction. Insulin concentration was determined by a monoclonal
antibody-based immunofluorimetric assay (AutoDelphia, Perkin-Elmer Life
Sciences Inc., Norton, OH, USA). Insulin resistance was calculated by the
homeostasis model assessment [HOMA-IR = fasting insulin (µU/mL)/22.5 x
fasting glycemia (mmoL/L)].
TSH
levels
were
measured
in
duplicate
by
a
sensitive
immunofluorimetric assay (AutoDELFIA, Wallac, Oy Turku, Finland) with a
reference range of 0.45 to 4.5 mU/L and a functional sensitivity of 0.05 mU/L.
Serum free thyroxine (FT4) was measured using a competitive immunoassay
(AutoDELFIA, Oy, Turku, Finland), wherein the normal reference range was
0.7–1.5 ng/dL.
78
Definitions
Euthyroidism was defined as a serum TSH level between 0.45 mU/L
and 4.5 mU/L, with an FT4 level within the normal reference ranges. SCH was
defined as a serum TSH level between 4.5 mU/L and 20 mU/L, with a normal
FT4 concentration.
Hypertension was defined as a blood pressure ≥ 140/90 mmHg or the
use of antihypertensive medication; diabetes was defined according to the
American Diabetes Association criteria; and dyslipidemia was defined by the
presence of any lipid abnormality (total cholesterol levels ≥ 200 mg/dL,
triglycerides ≥ 150 mg/dL, LDL-c > 130 mg/dL, or HDL-c < 40 mg/dL).
MS was defined according to the modified National Cholesterol
Education Program criteria (33) adapted for Asians (30), in which waist
circumference values of 102 cm for men and 88 cm for women were replaced
by 90 or 80 cm, respectively.
The traditional CV risk factors considered for analysis in this study were
smoking, hypertension, diabetes, and LDL-c, and the nontraditional CV risk
factors were waist circumference, triglycerides, apo B, HDL-c, CRP, IL-6,
homocysteine, uric acid, and HOMA-IR.
The diagnosis of cardiovascular disease at baseline was defined by a
medical history of congestive heart failure, angina pectoris, myocardial
infarction [confirmed by a physician and by major ECG abnormalities from old
infarctions (Q waves)], previous angioplasty, previous heart revascularization
procedure, or coronary insufficiency previously diagnosed by catheterization,
stroke, or intermittent claudication.
79
Statistical analysis
All statistical analyses were performed using SAS statistical software
version 9.1 (SAS Institute Inc., Cary, NC). The assumed level of significance
was p < 0.05 (two-tailed). Prevalence rates were calculated by point and
confidence interval (CI). The data were described using absolute (n) or relative
(%) frequencies, means with standard deviations (SD), and 95% confident
intervals (CI). Differences in the means of the characteristics between the
groups were assessed by Student’s t-test for parameters with a normal
distribution and by the Mann-Whitney U test for parameters without a normal
distribution. Frequencies were compared by the chi-squared test or Fisher’s
exact test when one of the absolute frequencies was below 5. Variables without
a normal distribution were subjected to logarithmic transformations before
statistical analysis. Multivariate logistic regression analysis, to obtain an odds
ratio (OR) and a 95% CI, was used to evaluate the association between SCH
and traditional and nontraditional CV risk factors, with and without adjustment
for age, smoking and body mass index (BMI). Spearman correlation analyses
were used to determine the correlation between the levels of serum TSH and
the studied variables. Based on expert reviews (1,5,6), a subset analysis was
further performed in a subgroup of individuals with SCH who had more severely
increased serum TSH levels (> 10 to 19.9 mU/L).
Results
The characteristics of the study population are shown in Table 1.
Participants included 913 individuals exhibiting euthyroidism (92.2%) and 99
(9.8%) SCH; the mean age was similar in the two groups. Compared with
80
euthyroid subjects, individuals with SCH were more frequently women (63.6%,
p = 0.02). As expected, the TSH levels of the SCH group were significantly
elevated (p < 0.0001) in both men and women, while the FT4 levels were
significantly lower (p < 0.001) in women, but not in men. The overall proportions
of smokers and patients with MS, diabetes, hypertension, dyslipidemia, and
cardiovascular disease were not significantly different between the groups. In
addition, there were no statistically significant differences concerning BMI, waist
circumference, systolic or diastolic blood pressure, fasting serum insulin,
HOMA-IR, total cholesterol, LDL-c, HDL-c, apo B, CRP, IL-6, or uric acid levels.
On the other hand, triglyceride (p = 0.009) and homocysteine (p = 0.04) levels
were both significantly higher in women with SCH, but not in men. Women with
SCH were also more likely to use statins than those with euthyroidism [OR, 5.8
(95% CI, 1.7–20.2)]. Because statin use could be masking a potential
association between serum levels of lipids and SCH in women, the analysis was
repeated excluding this condition, but the results did not change.
Table 2 demonstrates the traditional and nontraditional CV risk factors
in SCH individuals compared with the controls. A multivariate analysis showed
no association between SCH and the studied CV markers for either women or
men. Despite the significantly higher levels of triglycerides and homocysteine
found in women with SCH (Table 1), when hypertriglyceridemia (>150 mg/dL)
and hyperhomocysteinemia (>15 µmol/L) were defined, there were no
statistically significant differences in the proportions of the affected individuals
between the two groups with respect to gender (Table 2), and the ORs were not
significant.
81
Because MS represents a combination of multiple risk markers, we
examined the association between SCH and a cluster of both traditional and
nontraditional CV risks assessed in this study. The analysis showed no
significant association between SCH and a cluster of up to five risk factors after
adjustment for age, sex, smoking, and BMI (Table 3).
Because the progression of SCH to overt disease (1), adverse effects
from SCH (1), and mortality from SCH (15) are more likely associated with
serum TSH levels > 10 mU/L, we performed an analysis on the subset of SCH
individuals (n=13) with more severely increased serum TSH levels (10 to 19.9
mU/L). However, comparing this group with the euthyroid controls, we found no
statistically significant difference in the mean serum levels or in the proportion of
affected individuals related to the studied variables. Similarly, no significant ORs
for the risk of either traditional or nontraditional CV risk factors, or of a cluster of
these factors, were found between the two groups.
Finally, we found a significant positive correlation between TSH and
triglyceride levels (Spearman R = 0.0646, p = 0.04), and a significant negative
correlation between TSH and waist circumference (Spearman R = −0.0734, p =
0.02).
Discussion
In this study, conducted on a population of Japanese-Brazilians
characterized by the marked prevalence of MS (30), we found no association
between SCH and either traditional CV risk factors, such as smoking, diabetes,
hypertension, and increased LDL-c, or nontraditional CV risk factors, such as
waist circumference, HOMA-IR, low HDL-c, high CRP, high apo B,
high IL-6,
82
and hyperuricemia, or with a cluster of these metabolic factors. The only
positive findings were the higher concentrations of both triglycerides and
homocysteine in women (but not in men) with SCH compared with those
showing euthyroidism. However, the percentage of SCH women with both
hypertriglyceridemia and hyperhomocysteinemia did not differ from the control
group, and the ORs for these markers in SCH were not significant.
To the best of our knowledge, this is the first population-based study
that has explored the potential associations of SCH with the many
cardiovascular risk factors that cluster within MS in a population with a high
prevalence of metabolic disorders. A previous study demonstrated the impact of
cardiovascular causes of death in this population (34). Most recently, our group
also reported that SCH was significantly associated with mortality in this same
population in a 7.5-yr cohort study (11). Mortality was also significantly
associated with some components of MS in this cohort, but the causal role of
SCH in mortality remained significant, even after adjustments for these
prognostic factors.
It is has been recognized that both hypothyroidism and MS are
associated with similar metabolic changes, such as weight gain, mild
hypertension, dyslipidemia, and an increased risk for cardiovascular disease
(23,24). According to some recent studies, a possible link between these
conditions could be insulin resistance (21). In 117 women with type 2 diabetes,
there were strong positive associations between TSH levels and lipid
parameters with adverse cardiac risks at low insulin sensitivity (36). In a Dutch
population-based study (25), low-normal FT4 levels were significantly
associated with higher insulin resistance and with MS traits, and thyroid
83
hormone levels (FT4 and FT3 (free triiodothyronine)) were negatively correlated
with both total cholesterol and apo B. In contrast, we found no correlation
between either TSH or FT4 with fasting insulin, insulin resistance, or with
atherogenic lipoproteins, and no association was found between SCH and
these markers. The conflicting results might partially reflect differences in the
population characteristics, selection criteria, and different adjustments. In the
Dutch study (25), subjects who were not euthyroid were excluded, while in the
current study, individuals taking oral blood glucose-lowering medication were
not excluded, which might have limited our ability to detect potential
associations between SCH and insulin resistance.
In agreement with our findings,
hypothyroidism had no impact on
insulin sensitivity assessed with HOMA-IR in 22 totally thyroidectomized
patients (37), and no correlation was found between insulin levels and either
TSH or FT3 concentrations (38). In addition, TSH was not significantly
correlated with insulin sensitivity in a study using the gold standard
hyperinsulinemic euglycemic clamp technique, and the associations found
between TSH with both LDL-c and HDL-c were independent of insulin sensitivity
(39). Furthermore, in a recent study of a population from the Berlin/Potsdam
region, participants in the upper-normal TSH range just failed to be significantly
more insulin resistant (23). Interestingly, in the Dutch study (25), low-normal FT4
was also associated with higher triglycerides, lower HDL-c, and abdominal
obesity. However, these associations were independent of insulin resistance,
indicating that other mechanisms underlie the relation of FT 4 to these
components of MS. In summary, data about this issue are conflicting.
84
The present study is also in agreement with a Danish cohort in terms of
the association between SCH with higher concentrations of triglycerides (40). In
contrast with that study, we found no association between SCH and raised CRP
levels, a marker of inflammation and a strong predictor of cardiovascular risk
(28). Data from the NHANES survey (41) showed similar results, suggesting
that CRP does not appear to contribute to the increased risk for CHD in SCH
(1). IL-6, another marker of the proinflammatory state, was also not significantly
associated to SCH in this population (28). This result contrasts with a study in
which patients with SCH and Hashimoto thyroiditis showed higher CRP and IL-6
values. However, in that study, the inflammation might have been related to
autoimmune processes and not to thyroid status (42).
Hyperhomocysteinemia has also been considered as an independent
biomarker for progressive atherosclerosis (28), and thyroid hormone seems to
influence its plasma concentration (21). In this study, serum homocysteine
levels were significantly increased in women with SCH compared with controls,
but this finding is likely not related to the thyroid status because we found no
correlation between either TSH or FT4 with homocysteine, and the risk for this
marker was not significantly associated with SCH. Possible confounders could
be serum concentrations of folate, vitamin B12, or creatinine, which may be
linked to the pathogenesis of hyperhomocysteinemia in hypothyroid patients
(1,21). In recent reviews (1,21), no role for homocysteine as a marker of CHD in
SCH was observed.
No consistent association of SCH with lipid profiles (except for
hypertriglyceridemia in women) was found in this study. These findings agree
with a Japanese study (26) and previous large occidental population-based
85
studies (2), but differ from others (3,4,10). Recently, randomized and doubleblind studies of L-thyroxine and placebo found that SCH patients treated with Lthyroxine showed improved total cholesterol and LDL-c levels (17,43), but a
systematic review found only marginal evidence indicating an association
between thyroid hormone replacement and improvements in lipid profiles in
patients with SCH (44). Thus, there are no conclusive data addressing the
effects of SCH on lipid profiles.
Uric acid seems to cluster with MS and may be an independent CV risk
factor (45), but uric acid levels and the proportion of individuals with
hyperuricemia were both similar between the groups in this study. This is in
broad agreement with a recent population-based study (26), suggesting that
uric acid does not appear to be associated as a marker of CV risk in SCH.
Finally, the present study does not confirm an association of SCH with a
cluster of metabolic risk factors, as found in a previous Japanese study (26).
Despite some similarities between that study and the current one regarding
ethnicity and the prevalence rates of SCH, there are marked differences in
terms of age and lifestyle. In addition, that cohort was highly selective because
it included only survivors from the atomic bomb.
Our study has limitations. It is a cross-sectional study, implying that its
ability to infer causality is limited. In addition, the fact that participants taking
oral blood glucose-lowering medication were not excluded might have limited
our ability to detect potential associations between SCH and insulin resistance.
Another limitation is that the low number of individuals with TSH levels above 10
mU/L might have limited the power to detect statistical differences in this group.
86
In summary, SCH was not associated with either traditional or
nontraditional CV risk factors or with a cluster of these factors in a JapaneseBrazilian population with a high prevalence of MS. These findings suggest that
the increased risk of CHD events and CHD mortality associated with SCH (15)
might be caused by mechanisms other than insulin resistance and low-grade
cardiac inflammation.
References
1. Biondi B, Cooper DS 2008 The clinical significance of subclinical thyroid
dysfunction. Endocr Rev 29: 76 -131.
2. Tunbridge WM, Evered DC, Hall R, Appleton D, Brewis M, Clark F, Evans JG,
Young E, Bird T, Smith PA 1977 The spectrum of thyroid disease in a
community: the Whickham survey. Clin Endocrinol (Oxf) 7: 481- 493.
3. Canaris GJ, Manowitz NR, Mayor G, Ridgway EC 2000 The Colorado thyroid
disease prevalence study. Arch Intern Med 160: 526 - 534.
4. Hollowell JG, Staehling NW, Flanders WD, Hannon WH, Gunter EW, Spencer
CA, Braverman LE 2002 Serum TSH, T 4, and thyroid antibodies in the United
States population (1988 to 1994): National Health and Nutrition Examination
Survey (NHANES III). J Clin Endocrinol Metab 87: 489 - 499.
5. Helfand M 2004 Screening for subclinical thyroid dysfunction in nonpregnant
adults: a summary of the evidence for the U.S. Preventive Service Task Force.
Ann Intern Med 140:128–141.
87
6. Surks MI, Ortiz E, Daniels GH, Sawin CT, Col NF, Cobin RH, Franklyn JA,
Hershman JM, Burman KD, Denke MA, Gorman C, Cooper RS, Weissman NJ
2004 Subclinical thyroid disease: scientific review and guidelines for diagnosis
and management. JAMA 291: 228 - 238.
7. Gharib H, Tuttle M, Baskin HJ, Fish LH, Singer PA, McDermott MT
2005
Subclinical thyroid dysfunction: a joint statement on management from the
American Association of Clinical Endocrinologists, the American Thyroid
Association, and The Endocrine Society. J Clin Endocrinol Metab 90: 581 - 585.
8. Hak AE, Pols HA, Visser TJ, Drexhage HA, Hofman A, Witteman JC 2000
Subclinical hypothyroidism is an independent risk factor for atherosclerosis and
myocardial infarction in elderly women: the Rotterdam Study. Ann Intern Med
132: 270 - 278.
9. Rodondi N, Newman AB, Vittinghoff E, de Rekeneire N, Satterfield S, Harris TB,
Bauer DC 2005 Subclinical hypothyroidism and the risk of heart failure, other
cardiovascular events, and death. Arch Int Med 165: 2460-2466.
10. Walsh JP, Bremner AP, Bulsara MK, O'Leary P, Leedman PJ, Feddema P,
Michelangeli V 2005 Subclinical thyroid dysfunction as a risk factor for
cardiovascular disease. Arch Int Med 165: 2467-2472.
11. Sgarbi JA, Matsumura LK, Kasamatsu TS, Ferreira SR, Maciel RM
2010
Subclinical thyroid dysfunctions are independent risk factors for mortality in a 7.5year follow-up: the Japanese-Brazilian thyroid study. Eur J Endocrinol 162: 56977.
12. Ochs N, Auer R, Bauer DC, Nanchen D, Gussekloo J, Cornuz J, Rodondi N 2008
Meta-analysis: subclinical thyroid dysfunction and the risk for coronary heart
disease and mortality. Ann Int Med 148: 832-845.
88
13. Razvi S, Shakoor A, Vanderpump M, Weaver JU, Pearce SH 2008 The influence
of age on the relationship between subclinical hypothyroidism and ischemic heart
disease: a metaanalysis. J Clin Endocrinol Metabol 93: 2998-3007.
14. Volzke H, Schwahn C, Wallaschofski H, Dorr M 2007 Review: The association of
thyroid dysfunction with all-cause and circulatory mortality: is there a causal
relationship? J Clin Endocrinol Metabol 92: 2421-2429.
15. Rodondi N, den Elzen WP, Bauer DC, Cappola AR, Razvi S, Walsh JP, Asvold
BO, Iervasi G, Imaizumi M, Collet TH, Bremner A, Maisonneuve P, Sgarbi JA,
Khaw KT, Vanderpump MP, Newman AB, Cornuz J, Franklyn JA, Westendorp
RG, Vittinghoff E, Gussekloo J 2010 Thyroid Studies Collaboration. Subclinical
hypothyroidism and the risk of coronary heart disease and mortality. JAMA
304:1365-74.
16. Danese MD, Ladenson PW, Meinert CL, Powe NR 2000 Effect of thyroxine
therapy on serum lipoproteins in patients with mild thyroid failure: a quantitative
review of the literature. J Clin Endocrinol Metabol 85: 2993-3001.
17. Razvi S, Ingoe L, Keeka G, Oates C, McMillan C, Weaver JU 2007 The beneficial
effect of L-thyroxine on cardiovascular risk factors, endothelial function, and
quality of life in subclinical hypothyroidism: randomized, crossover trial. J Clin
Endocrinol Metabol 92: 1715-1723.
18. Cabral MD, Costa AJL, Santos M, Vaisman M 2004. Lipid profile alterations in
subclinical hypothyroidism. Endocrinologist 14:121-125.
19. Dagre AG, Lekakis JP, Protogerou AD, Douridas GN, Papaioannou TG,
Tryfonopoulos DJ, Papamichael CM, Alevizaki M 2007 Abnormal endothelial
89
function in female patients with hypothyroidism and borderline thyroid function.
Int J Cardiol 114: 332-338.
20. Chadarevian R, Bruckert E, Leenhardt L, Giral P, Ankri A, Turpin G 2001
Components of the fibrinolytic system are differently altered in moderate and
severe hypothyroidism. J Clin Endocrinol Metab 86:732-737.
21. Duntas LH, Wartofsky L 2007 Cardiovascular risk and subclinical hypothyroidism:
focus on lipids and new emerging risk factors. What is the evidence? Thyroid
17:1075-1084.
22. Uzunlulu M, Yorulmaz E, Oguz A 2007 Prevalence of subclinical hypothyroidism
in patients with metabolic syndrome. Endocr J 54 :71-76.
23. Ruhla S, Weickert MO, Arafat AM, Osterhoff M, Isken F, Spranger J, Schöfl C,
Pfeiffer AF, Möhlig M 2010 A high normal TSH is associated with the metabolic
syndrome. Clin Endocrinol (Oxf) 72 : 696-701.
24. Erdogan M, Canataroglu A, Ganidagli S, Kulaksızoglu M
2010 Metabolic
syndrome prevelance in subclinic and overt hypothyroid patients and the relation
among metabolic syndrome parameters. J Endocrinol Invest Jul 22. [Epub ahead
of print]
25. Roos A, Bakker SJ, Links TP, Gans RO, Wolffenbuttel BH 2007 Thyroid function
is associated with components of the metabolic syndrome in euthyroid subjects. J
Clin Endocrinol Metab 92:491- 496.
26. Ashizawa K, Imaizumi M, Usa T, Tominaga T, Sera N, Hida A, Ejima E, Neriishi K,
Soda M, Ichimaru S, Nakashima E, Fujiwara S, Maeda R, Nagataki S, Eguchi K,
Akahoshi M 2010 Metabolic cardiovascular disease risk factors and their clustering in
subclinical hypothyroidism. Clin Endocrinol (Oxf) 72: 689-695.
90
27. Grundy SM 2007 Metabolic syndrome: a multiplex cardiovascular risk factor. J
Clin Endocrinol Metab 92: 399-404.
28. Alberti KG, Zimmet P, Shaw J 2006 Metabolic syndrome--a new world-wide
definition. A Consensus Statement from the International Diabetes
Federation. Diabet Med 23: 469 – 480.
29. Gimeno SG, Ferreira SR, Franco LJ, Hirai AT, Matsumura L, Moisés RS 2002
Prevalence and 7-year incidence of Type II diabetes mellitus in a JapaneseBrazilian population: an alarming public health problem. Diabetologia 45: 16351638.
30. Rosenbaum P, Gimeno SG, Sanudo A, Franco LJ, Ferreira SR; JapaneseBrazilian Diabetes Study Group 2005 Analysis of criteria for metabolic syndrome
in a population-based study of Japanese-Brazilians. Diabetes Obes Metab 7: 352
- 359
31. Friedewald WT, Levy RJ, Fredrickson DS 1972 Estimation of the concentration of
low-density lipoprotein cholesterol in plasma, without use of the preparative
ultracentrifuge. Clin Chem 18: 499-502.
32. Pfeiffer CM, Huff DL, Gunter EW 1999 Rapid and accurate HPLC assay for
plasma total homocysteine and cysteine in a clinical laboratory setting. Clin
Chem 45: 290-292.
33. Executive Summary of The Third Report of The National Cholesterol Education
Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High
Blood Cholesterol In Adults (Adult Treatment Panel III). Expert Panel on
Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults 2001
JAMA285:2486-2497.
91
34. Massimino FC, Gimeno SG, Ferreira SR; Japanese-Brazilian Diabetes Study
Group 2007All-cause mortality among Japanese-Brazilians according to
nutritional characteristics Cad Saude Publica 23: 2145-2156.
35. Chubb SA, Davis WA, Davis TM 2005 Interactions among thyroid function, insulin
sensitivity, and serum lipid concentrations: the Fremantle diabetes study. J Clin
Endocrinol Metab 90: 5317-5320.
36. Owecki M, Nikisch E, Sowiński J 2006 Hypothyroidism has no impact on insulin
sensitivity assessed with HOMA-IR in totally thyroidectomized patients. Acta Clin
Belg 61: 69-73.
37. Owecki M, El Ali Z, Nikisch E, Sowiński J 2008 Serum insulin levels and the
degree of thyroid dysfunction in hypothyroid women Neuro Endocrinol Lett 29
:137-140.
38. Bakker SJ, ter Maaten JC, Popp-Snijders C, Slaets JP, Heine RJ, Gans RO 2001
The relationship between thyrotropin and low density lipoprotein cholesterol is
modified by insulin sensitivity in healthy euthyroid subjects. J Clin Endocrinol
Metab 86:1206-1211.
39. Kvetny J, Heldgaard PE, Bladbjerg EM, Gram J 2004 Subclinical hypothyroidism
is associated with a low-grade inflammation, increased triglyceride levels and
predicts cardiovascular disease in males below 50 years. Clin Endocrinol (Oxf)
61 :232-238.
40. Hueston WJ, King DE, Geesey ME 2005 Serum biomarkers for cardiovascular
inflammation in subclinical hypothyroidism. Clin Endocrinol (Oxf) 63: 582-587.
41. Taddei S, Caraccio N, Virdis A, Dardano A, Versari D, Ghiadoni L, Ferrannini E,
Salvetti A, Monzani F 2006 Low-grade systemic inflammation causes endothelial
92
dysfunction in patients with Hashimoto's thyroiditis. J Clin Endocrinol Metab 91:
5076-5082.
42. Cappola AR, Fried LP, Arnold AM, Danese MD, Kuller LH, Burke GL, Tracy RP,
Ladenson PW 2006 Thyroid status, cardiovascular risk, and mortality in older
adults. The JAMA 295: 1033-1041.
43. Teixeira PF, Reuters VS, Ferreira MM, Almeida CP, Reis FA, Melo BA, Buescu
A, Costa AJ, Vaisman M. Treatment of subclinical hypothyroidism reduces
atherogenic lipid levels in a placebo-controlled double-blind clinical trial. Horm
Metab Res. 2008; 40: 50-5.
44. Villar HCCE, Saconato H, Valente O, Atallah AN 2007 Thyroid hormone
replacement for subclinical hypothyroidism (Cochrane Review). In: The Cochrane
Library, Issue 3.
45. Baker JF, Krishnan E, Chen L, Schumacher HR 2005 Serum uric acid and
cardiovascular disease: recent developments, and where do they leave us? Am J
Med 118:816-826.
Acknowledgement
We are grateful to the Japanese-Brazilian population and to The JapaneseBrazilian Diabetes Study Group, particularly to Drs Amélia Hirai and Sueli
Gimeno. We thank Sirlei Siriane for the statistical assistance. We are also
grateful to Gilberto Furuzawa and Patricia Hiroka for technical assistance and to
Angela Faria for administrative assistance.
93
Table 1. Population baseline characteristics according to thyroid function status
Men
Controls
Number
Women
SCH
Controls
SCH
444
36
469
63
Mean age (yr)
56.5 ±12.6
58.9±12.9
56.2±12.2
58.2±12.0
TSH (mU/l)
1.58±0.92
6.68±2.49║
1.67±0.95
7.39±2.98║
FT4 (ng/dl)
1.09±0.18
1.08±0.19
1.04±0.16
0.97±0,2§
BMI (kg/m2)*
25.5±3.7
24.5±3.5
24.7±4.0
24.4±3.8
Waist circ.* (cm)
84.5±10.6
82.5±9.8
81.3±9.7
82.6±9.6
Systolic BP (mmHg)
135.4±23.0
135.0±21.6
130.3±25.4
132.6±27.1
Diastolic BP (mmHg)
81.6±13.3
82.2±14.4
77.3±12.9
75.9±13.7
Smokers, n (%)
86 (19.5)
4 (11.1)
34 (7.3)
6 (9.5)
Diabetes, n (%)
182 (41.0)
9 (25.0)
146 (31.1)
26 (41.3)
Hypertension, n (%)
184 (41.4)
18 (50.0)
158 (33.7)
25 (39.7)
Dyslipidaemia, n (%)
392 (88.3)
35 (97.2)
369 (78.7)
51(81.0)
7 (1.6)
-
6(1.3)
5 (7.9)‡
MS, n (%)
243 (54.7)
17 (47.2)
241(51.7)
31 (49.2)
CVD, n (%)
70 (15.8)
6 (16.6)
50(10.6)
8 (12.7)
FPG (mg/dL)
128.3±34.1
115.4±13.9
120.7±32.6
126.8±35.4
2h glucose (mg/dl)
171,7±85.7
136.3±45.4
161.6±67.3
176.0±102.4
Total Chol. (mg/dL)
214.1±41.7
210.4±53.1
215.9±40.5
216.1±44.4
LDL-c (mg/dL)
129.3±38.0
122.9±43.2
133.0±36.6
127.7±44.1
HDL-c (mg/dL)
49.6±11.7
47.9±14.4
51.9±10.0
52.2±11.7
269.2±227.1
257.0±215.5
196.6±136.8
249.4±189.6‡
Apo B,(mg/dL)
92.1±25.9
106.6±22.0
90.2±26.7
81.9±27.2
Homocysteine (µmol/L)
12.8±7.0
14.4±8.7
9.7±4.2
10.5±3.3†
CRP (mg/dL)
0.23±0.47
0.19±0.35
0.34±1.02
0.18±0.21
IL-6 (pg/mL)
1.07±0.64
0.98±0.34
1.04±1.3
1.08±1.02
Uric acid* (mg/dL)
7.1±1.8
7.3±2.2
5.4±1.3
5.4±1.3
HOMA-IR*
3.0±3.1
2.7±3.0
2.5±1.9
2.8±2.4
Statin usage, n (%)
Triglygerides (mg/dL)
94
Legend Table 1
Abbreviations: SCH, subclinical hypothyroidism; BMI, body mass index; Waist
circ., waist circumference; Systolic BP, systolic blood pressure; Diastolic BP,
diastolic blood pressure; MS, metabolic syndrome; CVD, cardiovascular
disease; FPG, fasting plasmatic glucose; Total chol., total cholesterol; LDL-C,
low- density lipoprotein-cholesterol; HDL-c, high-density lipoprotein cholesterol;
Apo-B, apolipoprotein B; CRP, C-reactive protein; IL-6, interleukin-6; Cr,
creatinine; HOMA-IR, homeostasis model assessment for insulin resistance.
Data are presented as mean ± SD, unless noted otherwise.
*, values were log transformed for statistical analysis
† p < 0.05; ‡ p< 0.01; § p < 0.001; ║ p < 0.0001
Table 2. Risk for traditional and nontraditional cardiovascular risk factors in subclinical
hypothyroidism individuals.
Men
Controls
n = 444
SCH
n = 36
Women
OR (95% CI)*
Controls
n = 469
SCH
n = 63
OR (95%CI)
95
Traditional CV risk factors
Smokers, n (%)
86 (19.5)
4 (11.1)
0.5 (0.2-1.5)
34 (7.3)
6 (9.5)
1.3 (0.5-3.3)
Diabetes, n (%)
182 (41.0)
9 (25)
0.5 (0.2-1.0)
146 (31.1)
26 (41.3)
1.6 (0.9-2.7)
Hypertension, n (%)
184 (41.4)
18 (50)
1.3 (0.7-2.7)
158 (33.7)
25 (39.7)
1.3 (0.7-2.2)
High LDL-c, n (%)
206 (46.4)
15 (41.7)
0.9 (0.4-1.7)
252 (53.7)
30 (47.6)
0.7 (0.4-1.2)
Hypertriglyceridaemia, n (%)
324 (74)
30 (83.3)
1.9 (0.8-4.9)
270 (58.4)
42 (67.7)
1.4 (0.8-2.5)
Low HDL-C, n (%)
72 (16.2)
13 (36.1)
0.7 (0.3-1.7)
42 (9.0)
5 (7.9)
1.1 (0.4-2.9)
High apo B, n (%)
305 (68.7)
25 (69.4)
0.8 (0.4-1.8)
335 (71.4)
44 (69.8)
1.0 (0.5-1.8)
High homocysteine, n (%)
74 (21.1)
8 (29.6)
1.5 (0.6-3.6)
26 (6.8)
3 (6.3)
0.9 (0.3-3.2)
High CRP, n (%)
183 (49.3)
12 (40)
0.7 (0.3-1.6)
228 (57.3)
25 (50.0)
0.7 (0.4-1.2)
High IL-6, n (%)
-
-
-
1 (0.7)
-
-
209 (47.1)
17 (47.2)
1.1 (0.5-2.1)
121 (25.8)
18 (28.6)
1.1 (0.6-1.9)
Nontraditional CV risk factors
Waist circumference (cm)
Hyperuricaemia, n (%)
Legend Table 2
Abbreviations: SCH, subclinical hypothyroidism; OR, odds ratio; CI, confident interval ;
LDL-C, low- density lipoprotein-cholesterol; HDL-c, high-density lipoprotein
cholesterol; apo-B, apolipoprotein B; CRP, C-reactive protein; IL-6, interleukin-6.
High LDL-c, ≥ 130 mg/dL; low HDL-c, < 40 mg/dL; high Apo-B, > 133 mg/dL;
hyperhomocysteinaemia, > 15 µmol/L; high CRP, > 0.11 mg/dL; high IL-6, > 9.7
pg/mL; hyperuricaemia, uric acid ≥ 7.0 mg/dL.
*, odds ratio adjusted for age, smoking, and BMI
Table 3. Risk for a cluster of cardiovascular risk factors in subclinical
hypothyroidism
Number of
CVrisk factors
Euthyroidism
SCH
OR* (95% CI)
96
(n = 913)
(n= 99)
0
8 (0.9)
1 (1.0)
1.0
1
76 (8.3)
10 (10.1)
1.05 (0.12 – 9.32)
2
161 (17.6)
17 (17.2)
0.84 (0.1 – 7.17)
3
194 (21.2)
28 (28.3)
1.15 (0.14 – 9.58)
4
181 (19.8)
18 (18.2)
0.8 (0.09 – 6.73)
5
167 (18.3)
10 (10.1)
0.48 (0.05 – 4.22)
Legend Table 3
Abbreviations: CVR, cardiovascular risk; SCH, subclinical hypothyroidism; OR,
odds ratio; CI, confidence interval
*, adjusted for age, sex, smoking, and BMI
97